CN106716995A - Stereoscopic video zooming and foreground and background detection in a video - Google Patents

Stereoscopic video zooming and foreground and background detection in a video Download PDF

Info

Publication number
CN106716995A
CN106716995A CN201580043225.8A CN201580043225A CN106716995A CN 106716995 A CN106716995 A CN 106716995A CN 201580043225 A CN201580043225 A CN 201580043225A CN 106716995 A CN106716995 A CN 106716995A
Authority
CN
China
Prior art keywords
frame
key element
modification
video
effect
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201580043225.8A
Other languages
Chinese (zh)
Inventor
贝赫鲁兹·A·马利基
萨尔文纳兹·萨克霍什
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bit An Nimeite Co Ltd
Original Assignee
Bit An Nimeite Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US14/301,146 external-priority patent/US9661309B2/en
Priority claimed from US14/301,143 external-priority patent/US9693042B2/en
Application filed by Bit An Nimeite Co Ltd filed Critical Bit An Nimeite Co Ltd
Publication of CN106716995A publication Critical patent/CN106716995A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/20—Image signal generators
    • H04N13/261—Image signal generators with monoscopic-to-stereoscopic image conversion
    • H04N13/264—Image signal generators with monoscopic-to-stereoscopic image conversion using the relative movement of objects in two video frames or fields
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/597—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding specially adapted for multi-view video sequence encoding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/20—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video object coding
    • H04N19/23—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video object coding with coding of regions that are present throughout a whole video segment, e.g. sprites, background or mosaic

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
  • Image Analysis (AREA)

Abstract

A method of analyzing a video may include determining at least one of a fastest moving element, a slow moving element and a dominant element within a video based on a first frame of the video and a second frame of the video. The method may also include determining at least one of a foreground and a background within the video based on at least one of the fastest moving element, the slow moving element and the dominant element.

Description

The detection of the foreground and background in the scaling and video of three-dimensional video-frequency
Technical field
Present disclosure is related to the foreground and background in the treatment of video data, including the scaling and video of three-dimensional video-frequency Detection.
Background technology
In recent years, three-dimensional (solid) imaging for generating 3 D video (for example, TV, film etc.) has become more next It is more popular.One reason for this is that having great improvement for creating the camera and post-production technology of 3 D video.It is three-dimensional The Another reason that video catches on is that the public of viewing amusement seems to be ready to pay extra-pay for this special-effect.
However, compared with two-dimentional (monoscopic) technology is used, three-dimensional technique for taking is caused to be shot using 3-D technology and regarded Frequency is substantially more expensive.It is having been made additionally, there are millions of portions but be not use 3-D technology shoot two-dimensional video.
Therefore, many people always strive to for two-dimensional video to be converted to 3 D video.However, for two-dimensional video to be converted to These methods of three-dimensional viewing are probably ineffective, are resource-intensive and/or fail to produce acceptable result (example Such as, cardboard is produced locally to cut out effect).
Claimed theme is not limited to the implementation for solving any shortcoming or only being operated in all such as above-mentioned environment Scheme.Conversely, this background be merely provided for explaination can put into practice one of some embodiments described herein it is exemplary Technical field.
The content of the invention
According to the one side of embodiment, a kind of method can include determining that first frame and monoscopic of monoscopic video Movement between second frame of video.The method can also be included based on mobile analysis camera effect, based on mobile and camera effect Fruit generation first changes frame and changes frame based on mobile and camera effect generation second.The method can also include being based on camera The left eye viewing frame of effect and at least one of the first modification frame and the second modification frame generation three-dimensional video-frequency.The method may be used also Seen with including the right eye based on camera effect and at least one of the first modification frame and the second modification frame generation three-dimensional video-frequency See frame.
According to the another aspect of embodiment, a kind of method of analysis video can include being based on the first frame of video and regard Second frame of frequency determines most fast mobile key element in video, move slowly at key element and Primary element (dominant element) At least one.The method can also be including at least in being based on most fast mobile key element, moving slowly at key element and Primary element It is individual to determine at least one of foreground and background in video.
The objects and advantages of embodiment will come real at least through the key element, feature and combination that are particularly pointed out in claim Now and complete.
It should be appreciated that foregoing general description and the following detailed description are all exemplary and explanat, and And do not limit the present invention for required protection.
Brief description of the drawings
Exemplary embodiment will be described and explained with extra characteristic and details by using accompanying drawing, in accompanying drawing：
Fig. 1 shows the example system for generating three-dimensional (3D) video based on monoscopic (2D) video；
Fig. 2 shows movement of the description based on key element between the different frame of monoscopic video to generate and monoscopic video The block diagram of associated modification frame；
Fig. 3 shows the block diagram for describing the modification frame that generation is associated with monoscopic video, wherein can be based on One or more of monoscopic video infer frame to generate one or more modification frames；
Fig. 4 shows and describes the left eye viewing based on the right translation effect generation three-dimensional video-frequency being associated with monoscopic video Frame and right eye watch the block diagram of frame；
Fig. 5 shows and describes the left eye viewing based on the left effect generation three-dimensional video-frequency being associated with monoscopic video Frame and right eye watch the block diagram of frame；
Fig. 6 is shown to describe and is seen based on the left eye for reducing camera effect generation three-dimensional video-frequency being associated with monoscopic video See that frame and right eye watch the block diagram of frame；
Fig. 7 is shown to describe and is seen based on the left eye for amplifying camera effect generation three-dimensional video-frequency being associated with monoscopic video See that frame and right eye watch the block diagram of frame；
Fig. 8 shows the example frame of monoscopic video, wherein the frame include most fast mobile key element, move slowly at key element and Primary element；
Fig. 9 be based on be associated with scene most move soon key element, move slowly in key element and Primary element at least one The flow chart of the illustrative methods of individual prospect and/or background to determine scene；
Figure 10 is shown for the movement based on the prospect and/or background being associated with monoscopic video from monoscopic video The left eye viewing frame and right eye for generating three-dimensional video-frequency watch the block diagram of frame；
Figure 11 be based on camera effect by monoscopic Video Quality Metric for three-dimensional video-frequency illustrative methods flow chart；
Figure 12 A show the example scenarios that can be perceived by left eye and right eye；
Figure 12 B depict example grid, it illustrates being likely to be present in the three-dimensional video-frequency that is associated with the scene of Figure 12 A Left eye viewing frame and right eye viewing frame in key element between exemplary offset；
Figure 12 C depict example grid, it illustrates unified multiplication factor is applied to Figure 12 B skew it Afterwards, the key element of Figure 12 B watches the skew of frame relative to its respective left eye viewing frame and right eye；
Figure 12 D depict example grid, it illustrates the unified summation factor is applied to Figure 12 B skew it Afterwards, the key element of Figure 12 B watches the skew of frame relative to its respective left eye viewing frame and right eye；
Figure 13 is the flow chart of the illustrative methods of the depth for adjusting three-dimensional video-frequency；And
Figure 14 is the flow chart of the illustrative methods of the focus for adjusting three-dimensional video-frequency, all these all according to described herein At least some embodiments are arranged.
Specific embodiment
The mankind have a binocular vision system, and the binocular vision system uses (about 6.5 lis of interval about two and half Rice) two eyes.Each eye watches the world from a slightly different visual angle.Brain is counted using the difference at these visual angles Calculate or measurement distance.This binocular vision system partly has the ability that object distance is determined with relatively good precision. The relative distance of the multiple objects in can determining visual field with the help of binocular vision.Can also be with the help of binocular vision Determine the relative distance of the multiple objects in visual field.
Three-dimensional (solid) imaging utilizes the depth perceived by binocular vision by the way that two sub-pictures are presented to beholder, wherein One sub-picture is presented to an eye (for example, left eye), and another sub-picture is presented to another eyes (for example, right eye).It is in Now substantially the same key element can be included to two images of eyes, but the key element in two sub-pictures can be offset from one another, To imitate the skew visual angle that can be in daily life perceived by the eyes of beholder.Therefore, beholder can perceive and be retouched by image The depth of the key element painted.
Traditionally, made approximately three to eight inches of two video sources (for example, camera) using the interval being mounted side by side 3 D video, to capture the scene that may be perceived by two different eyes.The distance is commonly known as " between centers " or " two Between " distance.Therefore, two video sources create two videos；One is used for left eye, and one is used for right eye.Two videos can be by Put together as three-dimensional (or " 3-D ") video, wherein right eye video is presented to the right eye of beholder, and left-eye video is presented to The left eye of beholder so that beholder perceives video in three dimensions.Conversely, according to some embodiments of present disclosure, standing Volumetric video can obtain three-dimensional video-frequency from the video obtained by using single (monoscopic) video source.Obtained from monoscopic video source To these videos can be referred to as " 2-D " or " monoscopic video ".
Term " video " can refer to the image of any mobile type, and can include such as, but not limited to film, TV Program, event (for example, competitive sports, concert etc.), family movie, the internet video of record etc..Video is by a series of " frame " (referred to as " frame " or " frame of video ") is constituted, and each frame shows can include the feelings of key element or object (being referred to as " key element ") The image of scape.Key element can be mobile or can be static.For example, frame can be including such as mountain, hills, tree The image of the landscape of the key elements such as wood, animal, building, aircraft, train, automobile.
According to some embodiments, first frame second frame (example relative to monoscopic video of the key element in monoscopic video Such as, the subsequent frame or previous frame of the first frame) between movement can be used for generate one or more modification first frames, its correspondence In the first frame and can be used for generate three-dimensional video-frequency.In some embodiments, the first frame and the second frame can be continuous Frame, and in other embodiments, one or more intermediate frames can be included between the first frame and the second frame.First frame and First frame of modification can include substantially the same key element, and the movement based on key element between the first frame and the second frame, the One or more elements in one frame and the first frame of modification are offset from one another.In addition, the movement of key element can be used for analysis relatively In the first frame and the second frame camera effect (for example, zooming effect, panning effect, rotates effe and/or static effect (stationary effect)).In some embodiments, the first of modification can be generated based on the camera effect of analysis Frame.
In some embodiments, can be in camera effect and the first frame and the first frame of modification based on determined by extremely The left eye that a few generation can be configured as being watched by the left eye of beholder watches frame.Similarly, it is also based on being determined Camera effect and the first frame and the generation of at least one of first frame of modification can be configured as being watched by the right eye of beholder Right eye viewing frame.The process can be repeated to multiple frames of monoscopic video, be seen with the corresponding left eye for generating three-dimensional video-frequency See frame and right eye viewing frame.Therefore, in such embodiments, can based on analyzed camera effect and the first frame and Movement between second frame associated there generates three-dimensional video-frequency from monoscopic video.
In addition, in some embodiments, the generation of left eye viewing frame and right eye viewing frame can be based on the first frame and The determination of the movement of at least one of the associated foreground and background of scene of monoscopic video of second frame.At these Or in other embodiments, can be based on being included in the scene that be associated with the first frame and the second frame most move soon key element, Key element and/or Primary element is moved slowly to determine prospect and/or background.
Furthermore it is possible to be included in the skew between the key element that left eye is watched in frame and right eye viewing frame by adjustment to change Depth value in three-dimensional video-frequency.In addition it is also possible to which key element may be in and be stood by adjusting the skew between key element Amendment is made in perception in the prospect or background of the associated scene of volumetric video.
Right eye viewing frame described herein and left eye viewing frame can also be referred to as " eye image " and " left-eye image ". In addition, right eye viewing frame and left eye viewing frame be used for any suitable 3D forms such as upper/lower, left/right,Hi-Fi 3D、3D or any other applicable 3D form generate three-dimensional video-frequency.
Fig. 1 illustrate some embodiments according to present disclosure for being generated based on monoscopic (2-D) video 101 The example system 100 of three-dimensional (3-D) video 103.System 100 can include three-dimensional (solid) video generation module 104 (hereafter Referred to as " three-dimensional video-frequency module 104 ").Three-dimensional video-frequency module 104 can include being configured as receiving monoscopic video 101 and inciting somebody to action single Visual field video 101 is converted to any suitable system, device or the equipment of three-dimensional video-frequency 103.For example, in some embodiments In, three-dimensional video-frequency module 104 can include being configured as making computing device for monoscopic video 101 to be converted into solid The software of the computer executable instructions of the operation of video 103.
Three-dimensional video-frequency module 104 can be configured as based on one or more elements between the frame of monoscopic video 101 Movement generate three-dimensional video-frequency 103.In some embodiments, three-dimensional video-frequency module 104 can be based on mobile generation modification Frame.The generation of the modification frame that can be performed by three-dimensional video-frequency module 104 further will in detail be begged for below in relation to Fig. 2 and Fig. 3 By.In some embodiments, three-dimensional video-frequency module 104 can be additionally configured to be analyzed based on key element movement between frames And determine the camera effect being associated with monoscopic video 101.In these embodiments, three-dimensional video-frequency module 104 can be by Be configured to identified camera effect and movement associated there generate for three-dimensional video-frequency 103 modification frame and Left eye watches frame and right eye viewing frame.Additionally, in some embodiments, three-dimensional video-frequency module 104 can be configured as being based on The determination of pair movement being associated with the foreground and background in the scene of monoscopic video 101 is seen generating modification frame and left eye See frame and right eye viewing frame.The analysis of camera effect and can based on analyzed camera effect and by three-dimensional video-frequency module The generation of the 104 modification frames for performing and left eye viewing frame and right eye viewing frame further will in detail be begged for reference to Fig. 4-7 and 10 By.
In some embodiments, three-dimensional video-frequency module 104 can be additionally configured to determine the scene of monoscopic video 101 Prospect and/or background.In some of these embodiments, three-dimensional video-frequency module 104 can be configured as being based on being included in Most fast mobile key element in the scene of monoscopic video 101, move slowly at key element, and/or Primary element determine prospect and/or Background.The determination of key element and/or Primary element will be described in further detail to most fast mobile key element, moved slowly on Fig. 8.Can The determination to prospect and/or background performed by three-dimensional video-frequency module 104 will be described in further detail below in relation to Fig. 9.
Additionally, in some embodiments, three-dimensional video-frequency module 104 can be configured to adjustment and be included in left eye sight Offset to adjust the depth perceived by the beholder of viewing three-dimensional video-frequency 103 between the key element seen in frame and right eye viewing frame Value.The modification to depth perception that can be performed by three-dimensional video-frequency module 104 will be made into one below in relation to Figure 12 A-12C and 13 Step is discussed in detail.In addition, in some embodiments, three-dimensional video-frequency module 104 can be additionally configured to adjust three-dimensional video-frequency 103 Focus so that three-dimensional video-frequency module 104 which key element for being perceived by beholder be able to may be in scene prospect or Amendment is made in perception in background.Being discussed in further detail below with reference to Figure 12 A, 12B, 12D and 14 can be by three-dimensional video-frequency The modification of the focusing that module 104 is performed.
As described above, three-dimensional video-frequency module 104 can be configured as based on one or more elements movement between frames To generate the modification frame that can be used for the left eye viewing frame for generating three-dimensional video-frequency 103 and/or right eye viewing frame.Fig. 2 shows basis Modification is described in the movement based on key element between the different frame 202 of monoscopic video 201 of some embodiments of the invention The block diagram 200 of the generation of frame 202'.In some embodiments, the embodiment for being given on Fig. 2, modification frame 202' can With by three-dimensional video-frequency module (all three-dimensional video-frequency modules 104 described as mentioned above for Fig. 1) generation.As shown in Fig. 2 monoscopic is regarded Frequently 201 the monoscopic video 101 on Fig. 1 descriptions can be substantially similar to, and series of frames 202, the frame 202 can be included The image of one or more scenes being associated with monoscopic video 201 can be included.According to some implementations of present disclosure Scheme, the water that three-dimensional video-frequency module can be based on the one or more elements (being not explicitly depicted) between frame 202 between frame 202 Translation is moved to generate modification frame 202'.
For example, key element (wherein frame 202b can be relative to the subsequent frame of frame 202a) in frame 202a and 202b can be There is change in location so that key element can from left to right be moved horizontally between frame 202a and 202b between frame 202a and 202b. Therefore, three-dimensional video-frequency module can move horizontally to generate modification frame 202a' based on key element between frame 202a and 202b, make Element must be wanted to be offset to the right relative to frame 202a in modification frame 202a'.Three-dimensional video-frequency module can be similarly based on frame 202b and 202c (wherein frame 202c can be relative to the subsequent frame of frame 202b) come generate modification frame 202b' and based on frame 202c and 202d (wherein frame 202d can be relative to the subsequent frame of frame 202c) changes frame 202c' to generate.Therefore, based on key element in frame Between move horizontally, the key element of modification frame 202' can offset relative to the corresponding factor level of correspondence frame 202, with imitate by The horizontal-shift of the image that the right eye and left eye of beholder are perceived.
In some embodiments, movement of the key element between frame 202 can also be (referred to as " vertical to move in vertical direction It is dynamic ").However, because the eyes of beholder do not offset generally substantially in vertical direction, as generation modification frame 202', Three-dimensional video-frequency module can substantially ignore key element vertically moving between frame 202.Therefore, it can generation modification frame 202', make Level is inclined in modification frame 202' based on moving horizontally relative to the corresponding key element in its associated frame 202 to obtain their key element Move.Even if however, vertically moving between frames, the key element of modification frame 202' is likely to will not be relative to its associated frame The 202 corresponding sizable degree of key element vertical shift.
In addition, in some embodiments, key element from the movement of a frame to another frame can cause frame in one or Multiple key elements are not present in the subsequent frame for can be used for the frame for generating modification.In this case, three-dimensional video-frequency module 104 can be with The key element lost is replicated from primitive frame and associated modification frame is placed them in.
For example, the key element of frame 202a can be the hand of people, and movement of the people between frame 202a and frame 202b can make The hand for obtaining people is invisible in frame 202b.Furthermore it is possible to be based on people move horizontally to generate modification from frame 202a to frame 202b Frame 202a'.Therefore, when moving horizontally generation modification frame 202a', can be based between frame 202a and 202b based on people The horizontal-shift of people horizontal-shift in modification frame 202a'.However, the hand of people can be replicated from frame 202a and be inserted into modification frame In 202a'.Therefore, even if hand may be lost from frame 202b, can be derived from changing the horizontal-shift of frame 202a' key elements, with In making to go smoothly and may reside in modification frame 202a'.
Without departing from the present invention, Fig. 2 can be modified, added or is omitted.For example, monoscopic is regarded Frequently 201 can include frame more more or less than the frame being explicitly illustrated.Additionally, in some embodiments, can be shown Include one or more intermediate frames between frame.For example, monoscopic video 201 can include can be in frame 202a and 202b, frame 202b And one or more intermediate frames between 202c, and/or frame 202c and 202d.In addition, in some embodiments, monoscopic is regarded Frequently the scene in 201 can change so that a frame 202 can be associated with a scene, and subsequent frame can be with difference Scene be associated.In the present invention, " scene " can refer to relative to the certain camera visual angle of certain scenarios or angle so that Change in " scene changes " or scene can refer to the change of the camera angle or visual angle of same scene and/or describe in video Scene change.In such embodiments, can be based on frame generation modification frame be inferred, wherein inferring that frame can be based on The movement of the one or more elements recognized in previous frame.
Fig. 3 shows the block diagram 300 of some embodiments according to present disclosure, which depict modification frame The generation of 302', wherein one or more that can be based on monoscopic video 301 infer the generation of frame 305, one or more change frames 302'.In some embodiments, and on Fig. 3 the embodiment for being given, modification frame 302' can be retouched by all as mentioned above for Fig. 1 The three-dimensional video-frequency module stated such as three-dimensional video-frequency module 104 is generated.As shown in figure 3, monoscopic video 301 can be including one Row frame 302 (is described as frame 302a-302d), and it can include one or more scenes being associated with monoscopic video 301 Image.In the illustrated embodiment, frame 302a-302c can be associated with the scene 306a of monoscopic video 301, and frame 302d can be associated with the different scenes (scene 306b) of monoscopic video 301.
According to some embodiments of present disclosure, three-dimensional video-frequency module can be configured as will based on one or more Plain (being not explicitly depicted) movement respectively between frame 302a and 302b and 302b and 302c come generate modification frame 302a' and 302b', similar to the generation of above-mentioned modification frame 202a' and 202b'.For example, can be based on one or more elements in frame 302a And the movement between frame 302b (it can be the subsequent frame of frame 302a) changes frame 302a' to generate.Furthermore it is possible to be based on one Or movement of multiple key elements between frame 302b and frame 302c (it can be the subsequent frame of frame 302b) generates modification frame 302b'。
However, in the illustrated embodiment, frame 302c can be associated most with the scene 306a of monoscopic video 301 Frame afterwards.Therefore, the movement of the key element being associated with frame 302c can be not present between the subsequent frame of frame 302c and frame 302c, because For the subsequent frame of such as frame 302d can be associated from different scenes, or can not have before another frame is transformed into The movement of their associated substantial extents.In all examples so, three-dimensional video-frequency module can be configurable to generate deduction Frame, infers that frame can represent that scene continues through the subsequent frame of the scene of its last frame.For example, in the illustrated embodiment, standing Stereoscopic frequency module can be configurable to generate deduction frame 305, infer that frame 305 can represent that scene 306a continues through frame 302c's The subsequent frame of the frame 302c of scene 306a, can be in the illustrated embodiment the last frame of scene 306a.
In some embodiments, three-dimensional video-frequency module can be configured to determine that scene changes, and can be based on inspection The scene changes for measuring determine the last frame of scene.Three-dimensional video-frequency module can be additionally configured to be based on detecting scene changes To generate deduction frame.For example, three-dimensional video-frequency module can be configured as detection from scene 306a to the scene changes of scene 306b, Can determine that frame 302c is the last frame of scene 306a, and can be based on detecting the scene from scene 306a to scene 306b Change infers frame 305 to generate.
In in these embodiments some, three-dimensional video-frequency module can be configured as being based on to the pixel between frame The analysis of space and/or distribution of color determines scene changes.If the distribution of pixel is from a frame to the basic phase of another frame Seemingly, then three-dimensional video-frequency module can determine that frame 302 is associated with identical scene 306.If however, the pixel distribution between frame Substantially different, then three-dimensional video-frequency module can determine to have occurred that scene change.For example, three-dimensional video-frequency module can determine frame Pixel distribution between 302a, 302b and 302c is substantially similar, to determine that frame 302a, 302b and 302c are probably scene 306a A part.Three-dimensional video-frequency module can also determine that the distribution of the pixel between frame 302c and 302d is substantially different, to determine in frame There is the scene change from scene 306a to scene 306b between 302c and 302d.
In response to detecting the scene change from the first scene to the second scene, three-dimensional video-frequency module can be configured as inspection Survey movement of the one or more elements between the frame of the first scene.Based on the movement between frame of the key element in the first scene, Three-dimensional video-frequency module can be configurable to generate the deduction frame of the first scene.When frame is inferred in generation, three-dimensional video-frequency module can be with The movement based on key element between the last frame of the first scene and deduction frame is configured as to generate the last frame of the first scene Modification frame.
For example, after detecting from scene 306a to the change of scene 306b, three-dimensional video-frequency module can be configured as Movement of the one or more elements between frame 302b and 302c is detected in response to the detection of scene changes.Based on what is detected Movement between frame 302b and 302c, three-dimensional video-frequency module can be configured as infer scene 306a continue through frame 302c it Afterwards, the movement that may occur between the subsequent frame in frame 302c and scene 306a.In some embodiments, can be based on as follows Assuming that carrying out the deduction of the movement between key element：If scene 306a does not terminate after frame 302c, key element is in frame 302b Mobile and 302c between will continue through frame 302c.Therefore, three-dimensional video-frequency module can be configured as based on key element in scene Movement between the frame 302b and 302c of 306a generates the deduction frame 305 of scene 306a so that infer that frame 305 can be reflected in The movement that scene 306a may continue when continuing through frame 302c in scene 306a.
Once generating deduction frame 305, three-dimensional video-frequency module can be based on one or more elements in frame 302c and deduction Movement between frame 305 changes frame 302c' to generate, and similar is based respectively between frame 302a and 302b and 302b and 302c Movement generate modification frame 302a' and 302b'.Therefore, it can based on may occurring to substantially from frame 302c The movement of the subsequent frame of identical element changes frame 302c' to generate, even if such subsequent frame may not exist.
Without departing from the present invention, Fig. 3 can be modified, added or is omitted.For example, monoscopic is regarded Frequently 301 can include frame more more or less than the frame being explicitly illustrated 302 and/or scene 306.In addition, each scene 306 can be with Including any amount of frame 302.Additionally, in some embodiments, one or more can be included between shown frame 302 Intermediate frame.
Fig. 1 is returned to, as it was previously stated, three-dimensional video-frequency module 104 can be configured as generating three-dimensional video-frequency based on camera effect 103 left eye viewing frame and right eye viewing frame, and the modification frame of monoscopic video 101 can be based on key element shifting between frames Move to generate, as explained above with described by Fig. 2 and Fig. 3.Therefore, one or more embodiments for being presented according to the present invention, Three-dimensional video-frequency module 104 can be additionally configured to analyze the camera effect of monoscopic video 101.In such embodiments, Three-dimensional video-frequency module 104 can determine that camera effect is panning effect or zooming effect, and can be based on camera effect Left eye watches frame and right eye watches frame to generate for panning effect or zooming effect.Further begged in detail below with reference to Fig. 4 and Fig. 5 By the determination left eye viewing frame and right eye viewing frame that generate that based on camera effect are panning effect.Enter referring to Fig. 6 and Fig. 7 One step is discussed in detail watches frame based on the determination generation left eye viewing frame and right eye that camera effect is zooming effect.
In addition, if three-dimensional video-frequency module 104 determines that camera effect is not panning effect or zooming effect, then three-dimensional video-frequency Module 104 can generate left eye viewing frame and right eye viewing frame according to the determination.In some of these embodiments, stereopsis The movement of the foreground and background that frequency module 104 can be associated based on the frame with monoscopic video 101 come generate left eye viewing frame Frame is watched with right eye.Foreground and background is more fully described relative to Fig. 8 and 9 to detect.Discussed in more detail below with reference to Figure 10 It is not the determination generation left eye viewing frame and right eye viewing frame of panning effect or zooming effect based on camera effect.
Fig. 4 show according to present disclosure it is some embodiments, depict the frame based on monoscopic video 401 To generate, the left eye of three-dimensional video-frequency 403 watches frame 410 to 402a and frame 402b associated right translation effect and right eye watches frame 412 Block diagram 400.In illustrated example, three-dimensional video-frequency 403 can be by three-dimensional video-frequency module, such as above with reference to Fig. 1 Described three-dimensional video-frequency module 104 is generated.
Frame 402a and 402b can include one or more elements 408, and these key elements 408 can be respectively included in frame In both 402a and frame 402b.For example, in the illustrated embodiment, frame 402a and 402b can each include key element 408a and 408b.Frame 402a and 402b can also be including other key elements not explicitly depicted in Fig. 4.
When camera is translated, by camera shoot whole scene with camera move and move so that frame in translation The associated key element of effect can be moved with substantially unified size and Orientation.For example, frame 402a and 402b can be flat with the right side Moving effect is associated, and wherein the camera of delta frame 402a and 402b may be translated from left to right so that frame 402a and 402b Interior key element 408a and 408b can from right to left be moved between frame 402a and 402b with substantially the same amount.
Right translation effect can be produced based on actual camera translation from left to right, or can be based on from left to right Analogue camera translates to produce.In some embodiments, panning effect can be with camera pivotable about an axis or its is any suitable Simulation be associated, to produce panning effect.In other embodiments, panning effect can be with whole camera in the horizontal direction It is associated to produce translation to imitate relative to the movement of the scene (for example, from left to right) being just taken or its any suitable simulation Really.
Three-dimensional video-frequency module can be configured as based on Essential Elements Of Analysis (such as key element 408a and 408b) in frame 402a and Movement between 402b determines to whether there is right translation effect between frame 402a and 402b.For example, in some embodiments In, three-dimensional video-frequency module can analyze the different zones positioned at frame 402a and 402b (for example, upper right, upper left, bottom right and lower-left area Domain) in different key elements movement.In some embodiments, three-dimensional video-frequency module can be by analysis and frame 402a and 402b The movement that may correspond to the pixel of key element is associated carrys out the movement of Essential Elements Of Analysis.In these or other embodiment, stereopsis Frequency module can analyze the movement being associated with each pixel of frame 402a and 402b, to determine key element in frame 402a and frame 402b Between movement.If with the key element (or being in some embodiments pixel) in the different zones of frame 402a and 402b Associated mobile whole is moved to largely the same extent from right to left, then three-dimensional video-frequency module can determine camera effect It is right translation effect.
Three-dimensional video-frequency module can be generated and frame 402a in right translation effect and movement associated there based on determined by Associated modification frame 402a'.For example, as described above, right translation effect can cause key element 408a and 408b from frame 402a to Frame 402b is moved from right to left to largely the same extent.Therefore, in some embodiments, because right translation effect causes The difference of the horizontal level of key element 408a and 408b between frame 402a and 402b, thus three-dimensional video-frequency module can use frame 402b changes frame 402a' to generate.In some embodiments, the key element 408a and 408b between frame 402a and 402b are little Or without in the case of vertical shift, frame 402b can serve as modification frame 402a'.Key element wherein between frame 402a and 402b Be there may be in 408a and 408b in other embodiments of vertical shift, can be removed during generation modification frame 402a' The upright position difference of key element 408a and 408b between frame 402a and 402b so that key element 402a and 402b is in frame 402a and repaiies Any vertical shift changed between frame 402a' can be substantial with right and wrong.
Left eye viewing frame 410 and the right side of the three-dimensional video-frequency 403 that can be associated with the viewing frame 402a of monoscopic video 401 Observe and see that frame 412 can be generated based on right translation effect, viewing frame 402a and modification viewing frame 402a'.For example, because the right side is flat Moving effect, frame 402a can be designated as left eye viewing frame 410, and change frame 402a' can be designated as right eye viewing frame 412.Therefore, it can be based at least partially on right translation effect and modification frame 402a'(can be with key element 408a and 408b in frame Movement between 402a and 402b is associated) to generate, the left eye of three-dimensional video-frequency 403 watches frame 410 and right eye watches frame 412.
In addition to being expressly recited, the generation of three-dimensional video-frequency 403 can be modified, added or omitted.For example, with The associated left eye viewing frame of other frames 402 of monoscopic video 401 and right eye viewing frame (for example, be associated with frame 402b Left eye watches frame and right eye viewing frame) can generate in a similar way.In addition, in some embodiments, can be with delta frame The left eye viewing frame and right eye viewing frame of 402 (can be the last frame of its corresponding scene or close to last frame) so that with frame 402 Associated modification frame can be generated based on the deduction frame determined above with respect to the mode described by Fig. 3.
Fig. 5 shows the block diagram 500 of some embodiments according to present disclosure, which depict based on list The frame 502a and frame 502b of visual field video 501 associated left effect watches frame 510 generating the left eye of three-dimensional video-frequency 503 Frame 512 is watched with right eye.In the example shown, three-dimensional video-frequency 503 can be (all to be retouched as explained above with Fig. 1 by three-dimensional video-frequency module The three-dimensional video-frequency module 104 stated) generation.
Frame 502a and 502b can include one or more that can be respectively included in both frame 502a and frame 502b Key element 508.For example, in the illustrated embodiment, frame 502a and 502b can each include key element 508a and 508b.
As described above, frame 502a and 502b can be associated with left effect, the wherein phase of delta frame 502a and 502b Machine may be translated from right to left so that the key element 508 in frame 502a and 502b can be between frame 502a and 502b with base Identical amount is from left to right uniformly moved in sheet.For example, in the illustrated embodiment, key element 508a and 508b can be based on Frame 502a and 502b associated left effect is moved from left to right between frame 502a and 502b with substantially the same amount.
Similar to right translation effect, left effect can be produced based on actual camera translation from right to left, or Can be produced based on analogue camera translation from right to left.Additionally, in some embodiments, panning effect can with around axle The camera of pivot or any appropriate simulation are associated, to produce panning effect.In other embodiments, panning effect can be with Whole camera is in the horizontal direction relative to the movement or any suitable simulation of the scene (for example, from right to left) being just taken It is associated to produce panning effect.
As discussed above in relation to Fig. 4, similar to right translation effect is determined whether there is, three-dimensional video-frequency module can quilt Analysis and key element (such as key element 508a and 508b, or be in some embodiments pixel) are configured in frame 502a and Between 502b be associated movement come determine between frame 502a and 502b whether there is left effect.If with positioned at frame Key element (or pixel in some embodiments) in the different zones of 502a and 502b it is associated it is mobile all from a left side to The right side is moved to largely the same extent, then three-dimensional video-frequency module can determine that camera effect is left effect.
Three-dimensional video-frequency module can be generated and frame 502a in left effect and movement associated there based on determined by Associated modification frame 502a'.For example, as described above, left effect can cause key element 508a and 508b from frame 502a to Frame 502b is moved from left to right to largely the same extent.Therefore, in some embodiments, three-dimensional video-frequency module can be with base In frame 502b generation modification frame 502a', reason is that left effect causes key element 508a and 508b between frame 502a and 502b Horizontal level difference.In some embodiments, the key element 508a and 508b between frame 502a and 502b almost without or In the case of not having vertical shift, frame 502b is used as modification frame 502a'.Key element wherein between frame 502a and 502b Be there may be in 508a and 508b in other embodiments of vertical shift, frame 502a can be removed during generation modification frame The upright position difference of key element 508a and 508b and 502b between so that the key element between frame 502a and modification frame 502a' Any vertical shift of 508a and 508b can be substantial with right and wrong.
The left eye viewing frame 510 and right eye of the three-dimensional video-frequency 503 being associated with the viewing frame 502a of monoscopic video 501 are seen See that frame 512 can be generated based on left effect, viewing frame 502a and modification viewing frame 502a'.For example, due to left Effect, modification frame 502a' can be designated as left eye viewing frame 510, and frame 502a can be designated as right eye viewing frame 512.Therefore, it can be based at least partially on left effect and modification frame 502a'(can be with key element 508a and 508b in frame Movement between 502a and 502b is associated) to generate, the left eye of three-dimensional video-frequency 503 watches frame 510 and right eye watches frame 512.
In addition to being expressly recited, the generation of three-dimensional video-frequency 503 can be modified, added or omitted.For example, with The associated left eye of other frames 502 of monoscopic video 501 and right eye viewing frame (for example, the left eye being associated with frame 502b and Right eye watches frame) can generate in a similar way.In addition, can (can be it with delta frame 502 in some embodiments The last frame of corresponding scene or close to last frame) left eye viewing frame and right eye viewing frame so that can be based on above with respect to Fig. 3 Deduction frame that described mode is generated generates the modification frame being associated with frame 502.
Fig. 6 shows the block diagram 600 of some embodiments according to present disclosure, which depict based on list The frame 602a and frame 602b of the visual field video 601 associated left eye viewing frame 610 for reducing camera effect generation three-dimensional video-frequency 603 Frame 612 is watched with right eye.It is illustrated it is exemplary in, three-dimensional video-frequency 603 can be by three-dimensional video-frequency module, such as above with reference to figure Three-dimensional video-frequency module 104 described by 1 is generated.
Frame 602a and 602b can include one or more that can be respectively included in both frame 602a and frame 602b Key element 608.For example, in the illustrated embodiment, frame 602a and 602b can each include key element 608a, 608b, 608c and 608d。
When camera reduces, the key element of the frame in generated by camera can be from a frame to another frame towards by frame-grab The center movement of scene.For example, frame 602a and 602b can be associated with the camera effect for reducing so that in frame 602a and 602b Key element 608a, 608b, 608c and 608d can be from frame 602a to frame 602b towards the scene captured by frame 602a and frame 602b Center movement, as shown in Figure 6.The camera effect of diminution can be based on actual camera and reduce generation, or can be based on mould The diminution of plan is produced.
Similar to right translation or left effect is determined whether there is, as discussed above for Fig. 4 and Fig. 5, stereopsis Frequency module can be configured to determine that between frame 602a and 602b, with the presence or absence of the camera effect for reducing, it is based on analysis and is located at frame Key element in the different zones (such as, upper left, upper right, lower-left and lower right area) of 602a and 602b (such as can respectively be located at frame Key element 608a, 608b, 608c and 608d in the upper left of 602a and 602b, upper right, lower-left and lower right area) in frame 602a and Movement between 602b.In some embodiments, three-dimensional video-frequency module can be by analysis and the frame 602a that may correspond to key element The movement associated with the pixel of 602b carrys out Essential Elements Of Analysis and moves.In these or other embodiment, three-dimensional video-frequency module can The movement that analysis is associated with each pixel of frame 602a and 602b, to determine shifting of the key element between frame 602a and frame 602b It is dynamic.In these or other embodiment, three-dimensional video-frequency module can analyze what is be associated with each pixel of frame 602a and 602b It is mobile, to determine movement of the key element between frame 602a and frame 602b.If with the different zones of frame 602a and 602b The associated movement of key element (or in some embodiments be pixel) generally toward the feelings captured by frame 602a and 602b The center of scape is moved, then three-dimensional video-frequency module can determine that camera effect is to reduce camera effect.
Three-dimensional video-frequency module can reduce camera effect and movement associated there to generate and frame based on determined by 602a associated modification frame 602a' and modification frame 602a ".For example, reduce camera effect can cause to be respectively included in frame Key element (for example, key element 608a and 608c) in the left side 614a and 614b of 602a and 602b is from frame 602a to frame 602b with basic Upper identical degree is moved from left to right.In addition, reduce camera effect can cause to be respectively included in frame 602a and 602b Key element (for example, key element 608b and 608d) in right side 616a and 616b from frame 602a to frame 602b to largely the same extent Move from right to left.
In some embodiments, three-dimensional video-frequency module can be based respectively on left side 614a of the key element in frame 602a and 602b Horizontal-shift and 614b between changes the left side 614c of frame 602a' to generate.In addition, during 614c on the left of generation, standing Stereoscopic frequency module can ignore any vertical shift of the key element between left side 614a and 614b.For example, key element 608a and 608c There can be the horizontal level substantially the same with the left side 614b of frame 602b in the left side 614c of modification frame 602a'. However, key element 608a and 608c can be with basic with the left side 614a of frame 602a in the left side 614c of modification frame 602a' Upper identical upright position.
In addition, three-dimensional video-frequency module can be based on the horizontal and vertical of the key element being included in the right side 616a of frame 602a The right side 616c of frame 602a' is changed to generate in position so that right side 616c can be substantially similar to right side 616a.For example, key element 608b and 608d can have substantially the same with the right side 616a of frame 602a in the right side 616c of modification frame 602a' Horizontal and vertical position.
Three-dimensional video-frequency module is also based on water of the key element respectively between the right side 616a and 616b of frame 602a and 602b Flat skew generates modification frame 602a " right side 616d.In addition, three-dimensional video-frequency module can be neglected during 616d on the right side of generation Omit any vertical shift of the key element between right side 616a and 616b.For example, key element 608b and 608d can be in modification frame 602a " Right side 616d in have the horizontal level substantially the same with the right side 616b of frame 602b.However, key element 608b and 608d can be in modification frame 602a " right side 616d in there is the vertical position substantially the same with the right side 616a of frame 602a Put.
By contrast, three-dimensional video-frequency module can be based on being included in key element in the left side 614a of frame 602a level and Upright position generates modification frame 602a " left side 614d so that left side 614d can be substantially similar to left side 614a.For example, Key element 608a and 608c can be in modification frame 602a " left side 614d in have and the substantially phase in the left side 614a of frame 602a Same horizontal and vertical position.
As shown in fig. 6, for modification the frame 602a' and 602a of above-mentioned diminution effect " configuration can cause key element 608a, 608b, 608c and 608d are in modification frame 602a' relative to modification frame 602a " horizontal-shift to the right.In addition, key element 608a, 608b, 608c and 608d can be in modification frame 602a' relative to modification frame 602a " have seldom even without vertical shift.
Left eye viewing frame 610 and the right side of the three-dimensional video-frequency 603 that can be associated with the viewing frame 602a of monoscopic video 601 Observe and see that frame 612 can be based on reducing camera effect and modification viewing frame 602a' and 602a " generate.For example, due to reducing phase Machine effect, modification frame 602a' can be designated as left eye viewing frame 610, and change frame 602a " right eye sight can be designated as See frame 612.Therefore, it can be based at least partially on diminution camera effect and modification frame 602a' and 602a " (can be with key element The movement of 608a, 608b, 608c and 608d between frame 602a and 602b be associated) come generate three-dimensional video-frequency 603 left eye see See frame 610 and right eye viewing frame 612.
In addition to being expressly recited, the generation of three-dimensional video-frequency 603 can be modified, added or omitted.For example, with Single left eye associated as other frames 602 of video 601 and right eye viewing frame are (for example, the left eye viewing being associated with frame 602b Frame and right eye viewing frame) can generate in a similar way.In addition, in some embodiments, can (can be with delta frame 602 The last frame of its corresponding scene or close to last frame) left eye viewing frame and right eye viewing frame so that with repairing that frame 602 is associated Changing frame can be generated based on the deduction frame generated above with respect to the mode described by Fig. 3.
Fig. 7 shows block diagram 700, which depict some embodiments and monoscopic according to present disclosure The left eye viewing frame 710 and right eye of the frame 702a and frame 702b of video 701 associated amplification camera effect watch the life of frame 712 Into.Frame 702a and 702b can include the one or more elements 708 being respectively included in frame 702a and frame 702b.For example, In the illustrated embodiment, frame 702a and 702b can each include key element 708a, 708b, 708c and 708d.
When camera amplifies, the key element of the frame in generated by camera can move away from capture from by a frame to another frame The center of scene.For example, as shown in fig. 7, frame 702a and 702b can with amplify camera effect be associated so that frame 702a with Key element 708a, 708b, 708c and 708d in 702b can move away from frame 702a and 702b capture from frame 702a to frame 702b Scene center.Amplifying camera effect can be based on actual camera amplification generation, or can be produced based on Simulation scale-up.
Similar to determining whether there is right translation, left or reducing camera effect, above for as described in Fig. 4-6, stood Stereoscopic frequency module can be configured as based on analysis positioned at frame 702a and 702b different zones (for example upper right, upper left, bottom right and Lower left region) in movement of the key element (such as key element 708a, 708b, 708c and 708d) between frame 702a and 702b determine With the presence or absence of amplification camera effect between frame 702a and 702b.In some embodiments, three-dimensional video-frequency module can pass through The movement that analysis is associated with the pixel that can correspond to key element of frame 702a and 702b carrys out the movement of Essential Elements Of Analysis.These or In other embodiments, three-dimensional video-frequency module can analyze the movement being associated with each pixel of frame 702a and 702b, with true Determine movement of the key element between frame 702a and frame 702b.If with the different zones of frame 702a and 702b key element (or It is in some embodiments pixel) associated movement all substantially moves away from the scene captured by frame 702a and 702b Center, then three-dimensional video-frequency module can determine camera effect be amplify camera effect.
Three-dimensional video-frequency module can be generated and frame in macrophotograph machine effect and movement associated there based on determined by 702a associated modification frame 702a' and modification frame 702a ".For example, amplify camera effect can cause to be respectively included in frame Key element (for example, key element 708a and 708c) in the left side 714a and 714b of 702a and 702b is from frame 702a to frame 702b from dextrad Move left substantially the same degree.In addition, amplify camera effect can cause to be respectively included in the right side of frame 702a and 702b Key element (for example, key element 708b and 708d) in 716a and 716b moves substantially the same from left to right from frame 702a to frame 702b Degree.
In some embodiments, three-dimensional video-frequency module can be based respectively on left side 714a of the key element in frame 702a and 702b Horizontal-shift and 714b between changes the left side 714c of frame 702a' to generate.In addition, three-dimensional video-frequency module can be on the left of generation Ignore any vertical shift of the key element between left side 714a and 714b during 714c.For example, key element 708a and 708c can be with There is the horizontal level substantially the same with the left side 714b of frame 702b in the left side 714c of modification frame 702a'.However, Key element 708a and 708c can have and the substantially phase in the left side 714a of frame 702a in the left side 714c of modification frame 702a' Same upright position.
In addition, three-dimensional video-frequency module can be based on being included in key element in the right side 716a of frame 702a horizontal level and The right side 716c of frame 702a' is changed to generate in upright position so that right side 716c can be substantially similar to the right 716a.For example, Key element 708b and 708d can have and the substantially phase in the right side 716a of frame 702a in the right side 716c of modification frame 702a' Same horizontal level and upright position.
Three-dimensional video-frequency module is also based on water of the key element respectively between the right side 716a and 716b of frame 702a and 702b Flat skew generates modification frame 702a " right side 716d.In addition, three-dimensional video-frequency module can be neglected during the generation of right side 716d Omit any vertical shift of the key element between right side 716a and 716b.For example, key element 708b and 708d can be in modification frame 702a " Right side 716d in have the horizontal level substantially the same with the right side 716b of frame 702b.However, key element 708b and 708d can be in modification frame 702a " right side 716d in there is the vertical position substantially the same with the right side 716a of frame 702a Put.
Additionally, three-dimensional video-frequency module can be based on the horizontal and vertical position of the key element being included in the left side 714a of frame 702a Put to generate modification frame 702a " left side 714d so that left side 714d can be substantially similar to left side 714a.For example, key element 708a and 708c can be in modification frame 702a " left side 714d in have it is substantially the same with the left side 714a of frame 702a Horizontal and vertical position.
As shown in fig. 7, for modification the frame 702a' and 702a of above-mentioned amplification effect " configuration can cause key element 708a, 708b, 708c and 708d are in modification frame 702a' relative to modification frame 702a " it is that level offsets to the left.In addition, key element 708a, 708b, 708c and 708d can be in modification frame 702a' relative to modification frame 702a " have seldom even without vertical Skew.
Left eye viewing frame 710 and the right side of the three-dimensional video-frequency 703 that can be associated with the viewing frame 702a of monoscopic video 701 Observe and see that frame 712 can be based on amplifying camera effect and modification viewing frame 702a' and 702a " generate.For example, due to amplifying phase Machine effect, modification frame 702a " can be designated as left eye viewing frame 710, and change frame 702a' and can be designated as right eye sight See frame 712.Therefore, it can be based at least partially on amplification camera effect and modification frame 702a' and 702a " (can be with key element The movement of 708a, 708b, 708c and 708d between frame 702a and 702b be associated) come generate three-dimensional video-frequency 703 left eye see See frame 710 and right eye viewing frame 712.
In addition to being expressly recited, the generation of three-dimensional video-frequency 703 can be modified, added or omitted.For example, with The associated left eye of other frames 702 of monoscopic video 701 and right eye viewing frame (for example, the left eye being associated with frame 702b and Right eye watches frame) can also generate in a similar way.In addition, in some embodiments, can (can be with delta frame 702 The last frame of its corresponding scene or close to last frame) left eye viewing frame and right eye viewing frame so that can based on frame 702 and more than Text generates the modification frame being associated with frame 702 on the corresponding deduction frame that mode described in Fig. 3 is generated.
Fig. 1 is returned to, three-dimensional video-frequency module 104 can be correspondingly configured to based on as above relative to flat described in Fig. 4-7 Move or zooming effect come generate three-dimensional video-frequency 103 left eye and right eye viewing frame.But, in some cases, camera effect can Can be somewhat different with translation or zooming effect.For example, camera effect can be rotating camera effect or stationary cameras effect.At this In the case of kind, three-dimensional video-frequency module 104 can determine that camera effect is not to translate or zooming effect, and can be according to the determination Generation left eye and right eye viewing frame.In some of these embodiments, three-dimensional video-frequency module 104 can be based on camera effect not Translation or the determination of zooming effect and the movement of the prospect and/or background that are associated based on the frame with monoscopic video 101 come Generation left eye viewing frame and right eye viewing frame.
Therefore, in some embodiments, three-dimensional video-frequency module 104 can be configured to determine that and monoscopic video 101 At least one of associated foreground and background so that three-dimensional video-frequency module can be analyzed to be associated with prospect and/or background Movement.In some of these embodiments, three-dimensional video-frequency module 104 can be configured as based on monoscopic video 101 Frame come determine to be associated with the frame of monoscopic video 101 most fast mobile key element, move slowly at key element and/or Primary element comes true Determine prospect and/or background.
Fig. 8 show the monoscopic video 801 of some embodiments according to present disclosure example frame 802a and 802b, wherein frame 802a and 802b include most fast mobile key element 820, move slowly at key element 822 and Primary element 824.Monoscopic Video 801 can be substantially similar to the monoscopic video 101 of Fig. 1.In the illustrated embodiment, frame 802a and 802b can be with list The same scene of visual field video 801 is associated, and frame 802b can be the subsequent frame of frame 802a.Three-dimensional video-frequency module, such as The three-dimensional video-frequency module 104 of Fig. 1 can be configured as the most fast mobile key element 820 of identification, move slowly at key element 822 and/or leading Key element 824.
Three-dimensional video-frequency module can be with analysis frame 802a and 802b determining most fast mobile key element 820, move slowly at key element 822 And/or Primary element 824.For example, three-dimensional video-frequency module can be with the key element in analysis frame 802b relative to frame 802a displacement, with Recognize most fast mobile key element 820 and move slowly at key element 822.Most fast mobile key element can be than moving at a slow speed from frame 802a to frame 802b Dynamic key element moves farther distance.Therefore, three-dimensional video-frequency module can Essential Elements Of Analysis from frame 802a to the displacement of frame 802b, and have The key element for having maximum displacement can be confirmed as most fast mobile key element.
In the illustrated embodiment, most fast mobile key element 820 can have from frame 802a to the maximum displacement of frame 802b, make Obtaining three-dimensional video-frequency module can be by most mobile key element 820 be identified as the most fast mobile key element from frame 802a to frame 802b soon.In addition, As shown in figure 8, move slowly at key element 822 can also be moved between frame 802a and frame 802b, but with than most fast mobile key element 820 smaller degree movements.Therefore, three-dimensional video-frequency module can will move slowly at key element 822 and be identified as than most fast mobile key element The slow key element of 820 movements.
In addition, three-dimensional video-frequency module can be configured to determine that Primary element 824 is that dominating in frame 802a and 802b will Element.In some embodiments, three-dimensional video-frequency module can be based on pair pixel being associated with frame 802a and 802b analysis come Key element in identification frame 802a and 802b so that the quantity of pixel being associated with specific factor can be determined.Therefore, at some In embodiment, three-dimensional video-frequency module can be configured as based on Primary element 824 and than being included in frame 802a and 802b The pixel of the substantial higher amount of other key elements is associated to recognize Primary element 824.Therefore, by analysis frame 802a and 802b, three-dimensional video-frequency module can be capable of identify that most fast mobile key element 820, move slowly at key element 822 and/or Primary element 824. As discussed in more detail below, the identification of these key elements can assist in the prospect and/or background being associated with frame 802a and 802b.
The diagram of Fig. 8 is used for the purpose of introducing most fast mobile key element, moves slowly at the concept of key element and Primary element behind. Therefore, without departing from the scope of this disclosure, Fig. 8 can be modified, added or is omitted.In some embodiment party In case, key element 820 is most moved soon and displacement of the key element 822 between frame 802a and 802b is moved slowly at can be based on most rapid traverse Move key element 820 and move slowly at the speed difference of key element 822 and change.For example, in some cases, most fast mobile key element 820 Can be moved at substantially the same speed with key element 822 is moved slowly at so that most fast to move key element 820 and move slowly at key element 822 displacement between frame 802a and 802b can be with substantially the same.In addition, in some embodiments, Primary element 824 can Moved between frame 802a and 802b, and in other embodiments, Primary element 824 can not be moved.Additionally, in certain situation Under, frame 802a and 802b can not include Primary element.Additionally, most fast move key element 820, move slowly at key element 822 and/or main The moving direction for leading key element 824 can change.
As set forth above, it is possible to the most fast mobile key element for being based on being associated with scene, move slowly at key element and/or Primary element To determine the prospect and/or background of scene.Fig. 9 is some embodiments according to present disclosure based on being associated with scene Most fast mobile key element, at least one of key element and Primary element is moved slowly to determine the prospect and/or background of scene The flow chart of illustrative methods 900.In some embodiments, method 900 can be by the three-dimensional video-frequency module 104 of such as Fig. 1 Three-dimensional video-frequency module implement.For example, three-dimensional video-frequency module 104 can be configured as computer instructions being used for performing It is determined that the operation of the prospect and/or background represented by one or more blocks of method 900.While shown as discrete area, but respectively Kind of block can be divided into extra block according to desired implementation method, be combined into less block or be eliminated.
Method 900 can start in frame 902, wherein can determine key element and/or pixel between the first frame and the second frame Movement.At frame 904, it may be determined that whether all key elements and/or pixel move in substantially the same direction.If institute There are key element and/or pixel to move in substantially the same direction, then method 900 may be advanced to frame 906.If all key elements And/or pixel is not moved in substantially the same direction, then method 900 may be advanced to frame 914.
In frame 906, it may be determined that the most fast mobile key element that is associated with the first frame and the second frame and move slowly at key element. In some embodiments, most fast mobile key element and slow can be determined relative to the displacement of the first frame based on the key element in the second frame The mobile key element of speed, such as above for the most fast mobile key element 820 between the frame 802a and 802b of Fig. 8 and move slowly at key element As 822 descriptions.
In frame 908, it may be determined that whether there is significant speed difference in most fast mobile key element and between moving slowly at key element. If there is significant speed difference, then method 900 may be advanced to frame 910.If without significant speed difference, method 900 May be advanced to frame 912.In some embodiments, speed difference can be by measuring most fast mobile key element and moving slowly at key element Determine to the difference of the side-play amount of the second frame from the first frame.In addition, in some embodiments, speed difference (for example, skew Difference) can be compared with threshold value, and if speed difference is more than threshold value, then method 900 may be advanced to frame 910, and if Speed difference is less than threshold value, then method 900 may be advanced to frame 912.In some cases, threshold value can be according to the type of film Change.For example, the threshold value of action movie can be higher than the threshold value of drama.
At frame 910, the speed difference between most moving key element soon and moving slowly at key element is substantially sufficiently large (for example, big In threshold value) in the case of, prospect can be related to most fast mobile key element so that most fast mobile key element is considered in prospect In.At frame 912, the speed difference between most moving key element soon and moving slowly at key element is not substantially sufficiently large (for example, small In threshold value) in the case of, background can be with related to the first frame and the associated whole scene of the second frame.Therefore, in frame 912 Place, the whole scene being associated with the first frame and the second frame is considered in the background.
Return at frame 904, if as described above, the movement of all key elements and/or pixel is differed, method 900 can To proceed to frame 914.At frame 914, it may be determined that wanted with the presence or absence of leading in the scene being associated with the first and second frames Element.In some embodiments, leading wanting can be determined whether there is based on the quantity of the pixel being associated with each key element The determination of element, all Primary elements 824 as mentioned above for described by Fig. 8.If there is Primary element, then method 900 can be in the past Enter frame 916.If there is no Primary element, then method 900 may be advanced to frame 922.
At frame 916, it may be determined that whether the Primary element determined in frame 914 lies substantially in and the first and second frames The center of associated scene.If Primary element is basic at center, method 900 may be advanced to frame 918.If leading will Plain basic Bu centers, then method 900 may be advanced to frame 920.
At frame 918, in the case of the basic center in scene of Primary element, prospect can be related to Primary element, makes It can be considered as prospect to obtain Primary element.At frame 920, the center of the scape that is absent from the scene substantially due to Primary element, background can be with Primary element is related so that Primary element is considered background.
Return at frame 914, as described above, if it is determined that in the absence of Primary element, then method 900 may be advanced to frame 922.At frame 922, it may be determined that the most fast mobile key element that is associated with the first and second frames and move slowly at key element, similar to Carried out in frame 906.At frame 924, it may be determined that whether there is in most fast mobile key element and between moving slowly at key element aobvious The speed difference of work, similar to what is carried out in frame 908.If there is significant speed difference, then method 900 may be advanced to frame 926.If there is no significant speed difference, then method 900 may be advanced to frame 928.
At frame 926, the substantially sufficiently large feelings of speed difference for most fast mobile key element and between moving slowly at key element Shape, prospect can be related to most fast mobile key element so that most fast mobile key element may be considered that in prospect.At frame 928, Speed difference for most fast mobile key element and between moving slowly at key element is substantially insufficient to big situation, and background or prospect can be with With related to the first frame and the associated whole scene of the second frame.Therefore, at frame 928, it is associated with the first frame and the second frame Whole scene may be considered that in background or prospect.
Therefore, method 900 is determined for the background and/or prospect of scene.It will be appreciated by those skilled in the art that right In this and other processes and method disclosed herein, can in a different order implement the work(performed in process and method Energy.Additionally, it is summarized the step of and operation only property is provided as an example, and some steps and operations can be optional, group The less step of synthesis and operation, or expand to additional step and operation, without departing from disclosed embodiment Essence.
Fig. 1 is returned to, as described above, in some embodiments, if three-dimensional video-frequency module 104 can determine that camera is imitated Fruit is not translation or zooming effect, then three-dimensional video-frequency module 104 can be based on the prospect being associated with the frame of monoscopic video 101 And/or the movement of background come generate left eye viewing frame and right eye viewing frame.
Figure 10 shows the block diagram 1000 according to some embodiments described herein, and it is based on being regarded with monoscopic Frequently the movement of 1001 associated prospects and/or background carrys out the left eye viewing from the generation three-dimensional video-frequency 1003 of monoscopic video 1001 Frame 1010 and right eye viewing frame 1012.Left eye watches frame 1010 and right eye viewing frame 1012 can be by three-dimensional video-frequency module, such as Three-dimensional video-frequency module 104 shown in Fig. 1 is generated.The left eye viewing frame 1010 and right eye viewing frame 1012 of three-dimensional video-frequency 1003 can Corresponding to the frame 1002a of monoscopic video 1001, and can be given birth to based on frame 1002a and its associated modification frame 1002a' Into.In the illustrated embodiment, can by it is all as described above with reference to Figure 2 in the way of based on movement between frame 1002a and 1002b To generate modification frame 1002a'；However, in other embodiments, can be in the way of above for described by Fig. 3 using pushing away Disconnected frame changes frame 1002a' to generate.
In the illustrated embodiment, three-dimensional video-frequency module is also based on the phase for determining to be associated with frame 1002a and 1002b Machine effect is not translation or zooming effect and based on related to frame 1002a and 1002b (and therefore also based on modification frame 1002a') To generate, left eye watches frame 1010 and right eye watches frame 1012 for the relative movement of the background and prospect of connection.In some embodiments In, three-dimensional video-frequency module can use the method 900 described above with reference to Fig. 9 to determine background and prospect.In other embodiment party In case, background and prospect can be in any other suitable manner determined.
For example, in some cases, prospect can be more mobile than background soon between frame 1002a and 1002b, and it is three-dimensional Video module can so determine.Which in this case, if prospect moves right, and ignore background can be in direction shifting Dynamic, frame 1002a can be appointed as left eye viewing frame 1010 by three-dimensional video-frequency module, and can be appointed as modification frame 1002a' Right eye watches frame 1012.In addition, in this case, if prospect is moved to the left, and ignore background may be at which Side is moved up, then modification frame 1002a' can be appointed as left eye viewing frame 1010 by three-dimensional video-frequency module, and can be by frame 1002a is appointed as right eye viewing frame 1012.
In other cases, background can be more mobile than prospect soon between frame 1002a and 1002b, and three-dimensional video-frequency Module can so determine.Which in this case, if background is moved to the left, and ignore prospect can be in direction shifting Dynamic, then frame 1002a can be appointed as left eye viewing frame 1010 by three-dimensional video-frequency module, and can be specified modification frame 1002a' For right eye watches frame 1012.In addition, in this case, if background moves right, and ignoring prospect can be in which side Move up, then three-dimensional video-frequency module can will change frame 1002a' and be appointed as left eye viewing frame and can refer to frame 1002a It is set to right eye viewing frame 1012.
Therefore, three-dimensional video-frequency module can be configured as be not according to the camera effect that is associated with frame 1002a and 1002b Translation or the determination of zooming effect, based on the back of the body being associated with frame 1002a and 1002b (therefore being also based on changing frame 1002a') To generate, left eye watches frame 1010 and right eye watches frame 1012 for the relative movement of scape and prospect.
Without departing from the scope of this disclosure, Figure 10 can be modified, added or omitted.For example, such as It is upper described, in the illustrated embodiment, frame 1002b can after frame 1002a, and can based on frame 1002a and 1002b it Between movement generate modification frame 1002a'.However, in some cases, if last frame such as special scenes or connect Nearly last frame generation modification frame, then the modification of such case can be generated based on deduction frame as described above with reference to Figure 3 Frame.In addition, in some embodiments, method 900 that can be according to Fig. 9 detects foreground and background, and at other In embodiment, foreground and background can be detected at other acceptable modes with certain.
The foregoing description of Fig. 1-10 shows that three-dimensional video-frequency module is (for example, figure according to some embodiments described herein Three-dimensional video-frequency module 104 shown in 1) movement that can be configured as between the frame based on monoscopic video and it is identified with The associated camera effect of monoscopic video generates left eye and right eye the viewing frame of three-dimensional video-frequency.Figure 11 is according in the disclosure Hold some embodiments based on camera effect by monoscopic Video Quality Metric for three-dimensional video-frequency illustrative methods 1100 stream Cheng Tu.In some embodiments, method 1100 can be come by the three-dimensional video-frequency module of the three-dimensional video-frequency module 104 of such as Fig. 1 Implement.For example, three-dimensional video-frequency module 104 can be configured as computer instructions to perform by one or many of method 1100 Individual frame represent for by monoscopic Video Quality Metric for three-dimensional video-frequency operation.While shown as discrete frame, but according to the phase The implementation of prestige, each frame can be divided into additional frame, be combined into less frame or be eliminated.
Method 1100 can start, and at frame 1102, it may be determined that first frame and monoscopic of monoscopic video are regarded Movement between second frame of frequency.At frame 1104, the first frame of movement generation modification can be based on.In some embodiments In, the first frame of modification can be generated based on the description be given with reference to Fig. 2.In other embodiments, can be based on reference Deduction frame described by Fig. 3 generates the first frame of modification.
At frame 1106, camera effect can be analyzed based on the movement and determined.For example, camera effect can be determined For panning effect, zooming effect or it is not.At frame 1108, camera effect analysis and the first frame can be based on modification At least one of first frame come generate three-dimensional video-frequency left eye watch frame.At frame 1110, camera effect analysis can be based on And first at least one of first frame of frame and modification watch frame generating the right eye of three-dimensional video-frequency.In the feelings of zooming effect Under condition, in some embodiments, the first frame of the first frame and the modification of the second frame generation second is also based on, and can be with base At least one of the first frame and the second the first frame changed in the first modification generation left eye viewing frame and right eye viewing frame. In some embodiments, can be according to the description generation left eye viewing frame and right eye viewing frame be given in Fig. 4-10.
Therefore, it is three-dimensional video-frequency that method 1100 can be used for monoscopic Video Quality Metric.Those skilled in the art should manage Solution, for this and other processes and method disclosed herein, can implement to be performed in process and method in a different order Function.Additionally, it is summarized the step of and operation only property is provided as an example, and some steps and operations can be optional , be combined into less step and operation, or expand to additional step and operation in, without departing from disclosed embodiment party The essence of case.
For example, in some embodiments, method 1100 can include and determine the step that background and/or prospect are associated Suddenly, the background and/or prospect can then be used to generate three-dimensional video-frequency, all as explained above with described by Fig. 9 and Figure 10. In other embodiments, foreground and background can determine in another way, or can be specified by input.
Fig. 1 is returned to, as described above, three-dimensional video-frequency module 104 can be configured to adjustment is included in left eye Offset to adjust the amount of the depth being associated with three-dimensional video-frequency 103 between key element in viewing frame and right eye viewing frame.Also such as Upper described, in some embodiments, three-dimensional video-frequency module 104 can be additionally configured to adjust the focus of three-dimensional video-frequency 103, make Three-dimensional video-frequency module 104 can may be in sense in the prospect of scene or background to which key element for being perceived by beholder Know and make amendment.The adjustment of depth and the adjustment of focus are described referring to Figure 12 A to Figure 14.
Figure 12 A show the example scenarios that can be perceived by left eye 1202 and right eye 1204 according to some embodiments 1200.Scene 1200 can include nearside key element 1206, intermediate elements 1208 and distally key element 1210.In example shown, Left eye 1202 and right eye 1204 can be focused in intermediate elements 1208 so that intermediate elements 1208 can be on left eye 1202 and the right side The focal point of eye 1204.Therefore, nearside key element 1206 can be in the prospect for such as being perceived by left eye 1202 and right eye 1204.Separately Outward, because left eye 1202 and right eye 1204 are focused in intermediate elements 1208, then distally key element 1210 can be such as by left eye 1202 and the background that perceives of right eye 1204 in.
Figure 12 B depict the example grid 1201 according to some embodiments disclosed herein, it illustrates may deposit It is the left eye viewing frame of the three-dimensional video-frequency being associated with scene 1200 and the key element 1206,1208 and 1210 of right eye viewing frame in Between exemplary offset.As described above, the skew of key element 1206,1208 and 1210 can simulate left eye 1202 and right eye 1204 Relative to the different visual angles of scene 1200.
In the illustrated embodiment, left side nearside key element 1206a can represent figure from the visual angle of the left eye 1202 of Figure 12 A The nearside key element 1206 of 12A, and the right nearside key element 1206b can represent Figure 12 A's from the visual angle of the right eye 1204 of Figure 12 A Nearside key element 1206.Therefore, alternate position spikes of the left side nearside key element 1206a relative to the right nearside key element 1206b in grid 1201 The nearside key element 1206 that the different left eye that can represent associated three-dimensional video-frequency is watched in frame watches the skew of frame relative to right eye.
In the illustrated embodiment, left side nearside key element 1206a can on the right, because as illustrated in fig. 12, nearside key element 1206 can be on the right for the sight line 1212 being associated with left eye 1202.In addition, the right nearside key element 1206b can on the left side, Because nearside key element 1206 can be on the left side of the sight line 1214 being associated with right eye 1204.This phenomenon is commonly known as " negative Parallax ", and occur in the case of target is before two focuses of eyes, in the illustrated embodiment, the target can To be set to intermediate elements 1208.
As viewed from the perspective of the left eye 1202 and right eye 1204 of Figure 12 A, the intermediate elements 1208a/b of grid 1201 can be with table The intermediate elements 1208 of diagram 12A.In the embodiment shown in Figure 12 A, intermediate elements 1208 can be on left eye 1202 and the right side The focal point of eye 1204 so that there may be between the visual angle of left eye 1202 and right eye 1204 seldom even without skew.Cause This, intermediate elements 1208 can be substantially the same in the right eye viewing frame with associated three-dimensional video-frequency and left eye viewing frame Position.Therefore, the intermediate elements 1208a/b of Figure 12 B show the left side intermediate elements 1208a that is perceived by left eye 1202 and by The overlap of right eye 1204 is perceived the right intermediate elements 1208b.
In addition, left side distally key element 1210a and the right distally key element 1210b can respectively from Figure 12 A left eye 1202 and the right side The visual angle of eye 1204 represents the distally key element 1210 of Figure 12 A.Therefore, the left side distally key element 1210a in grid 1201 relative to The position difference of the right distally key element 1210b can represent the distally key element in the left eye viewing frame of associated three-dimensional video-frequency 1210 skews that frame is watched relative to right eye.
In the illustrated embodiment, left side distally key element 1210a can be on the left side, because as illustrated in fig. 12, distally key element 1210 can be on the left side of the sight line 1212 being associated with left eye 1202.Additionally, the right distally key element 1210b can on the right, Because distally key element 1210 can be on the right for the sight line 1214 being associated with right eye 1204.The phenomenon is commonly known as " to be faced Difference " and generation in the case of target is two focuses of eyes behind, in the illustrated embodiment, the target can be At intermediate elements 1208.
Can be adjusted by adjusting the skew between left eye viewing frame and the corresponding key element of its associated right eye viewing frame The amount of the depth of whole three-dimensional video-frequency.For example, the amount of the depth being associated with the three-dimensional video-frequency being associated with Figure 12 B can be by adjusting Skew between whole left side nearside key element 1206a and the right nearside key element 1206b, by adjust left side intermediate elements 1208a and Skew between the right intermediate elements 1208b, and/or by adjusting left side distally key element 1210a and the right distally key element 1210b Between skew adjust.In some embodiments of present disclosure, can be united by the offset applications between key element One multiplication factor carrys out percentage regulation.
Figure 12 C depict the example grid 1203 according to some embodiments disclosed herein, it illustrates will system One multiplication factor is applied to after the skew of Figure 12 B, and the key element of Figure 12 B is relative to their own left eye and right eye viewing frame Skew.Unified multiplication factor can apply to the essentially all of key element phase watched in frame and right eye viewing frame with left eye The skew of association so that the skew between key element is adjusted by substantially the same yardstick.In the illustrated embodiment, it is unified Multiplication factor can have be 2 value, and can apply to the key element 1206 of Figure 12 B, the skew between 1208 and 1210, So that key element 1206 in Figure 12 C, the skew between 1208 and 1210 can each relative to Figure 12 B in corresponding skew add Times.
For example, in Figure 12 B, between the center of the center of left side nearside key element 1206a and the right nearside key element 1206b The skew of nearside key element can be about " 2 " individual grid cell.Therefore, nearside key element offset applications multiplication factor " 2 " can be caused Nearside key element offset approximation in Figure 12 C between the center of the center of left side nearside key element 1206a and the right nearside key element 1206b It is " 4 " individual grid cell, as indicated in fig. 12 c.
In addition, the intermediate elements skew between left side intermediate elements 1208a and the right intermediate elements 1208b can be about " 0 " individual grid cell.Therefore, multiplication factor " 2 " being applied into intermediate elements skew can cause the left side centre in Figure 12 C to be wanted Intermediate elements skew between plain 1208a and the right intermediate elements 1208b is still about " 0 " individual grid cell, because 1 number It is still " 0 " that word is multiplied by " 0 ".
Additionally, in Figure 12 B, between the center of the center of left side distally key element 1210a and the right distally key element 1210b The skew of distally key element can be about " 3 " individual grid cell.Therefore, distally key element offset applications multiplication factor " 2 " can be led Cause the distally key element skew between the center of left side distally key element 1210a in Figure 12 C and the center of the right distally key element 1210b " 6 " individual grid cell is approximately, as indicated in fig. 12 c.
In the illustrated embodiment, the right nearside key element 1206b being associated with right eye viewing frame can be big with shifted left About " 2 " individual grid cell, and the left side nearside key element 1206a being associated with left eye viewing frame can in fig. 12 c relative to Figure 12 B are not shifted.In addition, in the illustrated embodiment, the right distally key element 1210b being associated with right eye viewing frame can be to Dextroposition " 3 " individual grid cell, and the left side distally key element 1210a being associated with left eye viewing frame can phase in fig. 12 c Do not shifted for Figure 12 B.Therefore, watch frame by adjusting right eye watches frame without adjustment left eye, and the skew of nearside key element can be with Increase to " 4 " individual grid cell from " 2 " individual grid cell relative to Figure 12 B in fig. 12 c, and the skew of distally key element can be from " 3 " individual grid cell increases to " 6 " individual grid cell.
In an alternative embodiment, left side nearside key element 1206a and the left side of frame in can be watched by shifting left eye Distally key element 1210a is adjusted rather than the right nearside key element 1206b and the right distally key element 1210b in displacement right eye viewing frame The whole skew of nearside key element and the skew of distally key element in fig. 12 c relative to Figure 12 B.In other embodiments, can pass through Both displacement left side nearside key element 1206a and the right nearside key element 1206b adjust nearside in fig. 12 c relative to Figure 12 B Key element offsets.Additionally, in these or some embodiments of other embodiments, can be by shifting left side distally key element Both 1210a and the right distally key element 1210b adjust distally key element skew in fig. 12 c relative to Figure 12 B.
Although can be using any suitable in the above-described embodiments using specific unified multiplication factor " 2 " Multiplication factor.It is, for example possible to use increasing the depth perceived by beholder more than any appropriate multiplication factor of " 1 ".Separately Outward, can be used for reducing the depth perceived by beholder less than any suitable multiplication factor of " 1 ".Although additionally, above will Unified multiplication factor is described as the skew for being applied to be associated with corresponding key element, but in some embodiments, the key element Can each include one or more pixels, and unified multiplication factor can be applied to the skew that is associated with respective pixel.One In a little embodiments, unified multiplication factor can apply to each skew being associated with each key element and/or pixel.
Therefore, according to some embodiments of present disclosure, three-dimensional video-frequency module (for example, three-dimensional video-frequency module 104) Can be configured to be applied to unified multiplication factor to be watched with the left eye viewing frame of three-dimensional video-frequency and corresponding right eye Offset to adjust the depth being associated with three-dimensional video-frequency between the associated corresponding key element of frame and/or pixel.By contrast, Traditional depth adjustment programme may not apply the skew of common scale to carry out percentage regulation.
Can also be watched by adjusting left eye skew that frame and its associated right eye watched between the corresponding key element of frame come Adjust the focus of three-dimensional video-frequency.For example, can be by adjusting between left side nearside key element 1206a and the right nearside key element 1206b Skew, by adjusting the skew between left side distally key element 1210a and the right distally key element 1210b and by adjusting the left side Skew between intermediate elements 1208a and the right intermediate elements 1208b adjusts Jiao of the three-dimensional video-frequency being associated with Figure 12 B Point.In some embodiments, focus can be adjusted by the unified summation factor of the offset applications between key element.It is unified The summation factor can apply to be watched with left eye the skew that is associated of essentially all of key element of frame and right eye viewing frame in. In addition, in some embodiments, the unified summation factor can apply to be watched with left eye and right eye the respective pixel phase of frame The skew of association.
Figure 12 D depict the example grid 1205 according to some embodiments disclosed herein, it illustrates will system The one summation factor is applied to after the skew of Figure 12 B, and the key element of Figure 12 B is relative to their own left eye and right eye viewing frame Skew.In the illustrated embodiment, the unified summation factor can be applied so that the key element phase being associated with right eye viewing frame For the key element shifted left being associated with left eye viewing frame.This right eye viewing frame key element relative to left eye viewing frame key element to Shifting left can cause focus to be moved rearwards by so that more key elements are perceived in prospect.
In other embodiments, the unified summation factor can be applied so that the key element being associated with right eye viewing frame Relative to the key element right shift being associated with left eye viewing frame.This right eye viewing frame key element watches frame key element relative to left eye Right shift can cause focus to move forward so that more key elements are perceived in background.
In the embodiment illustrated, the unified summation factor can have value " -3 ", and can apply to Figure 12 B's Key element 1206, the skew between 1208 and 1210 so that each of right eye viewing frame key element can be relative to their corresponding left sides Observe and see three grid cells of frame key element shifted left.In the embodiment illustrated, bearing the summation factor can cause right eye to be seen See frame key element relative to left eye watch frame key element shifted left, and the factor of just suing for peace can cause right eye watch frame key element relative to Left eye watches frame key element right shift.However, in other embodiments, the positive summation factor can cause right eye to watch frame key element Frame key element shifted left is watched relative to left eye, and the negative summation factor can cause right eye to watch frame key element relative to left eye viewing Frame key element right shift.
For example, in Figure 12 B, between the center of the center of left side nearside key element 1206a and the right nearside key element 1206b The skew of nearside key element can be " 2 " individual grid cell.Therefore, the summation factor of " -3 " is applied to the nearside shown in Figure 12 B Key element skew can cause the right nearside key element 1206b relative to the individual grid list in left side nearside key element 1206a shifted lefts " 3 " Unit so that the nearside skew between the center of the center of left side nearside key element 1206a and the right nearside key element 1206b is in application Can be " 5 " individual grid cell after the summation factor, as indicated in fig. 12d.
In addition, the intermediate elements skew between left side intermediate elements 1208a and the right intermediate elements 1208b can be about " 0 " individual grid cell.Therefore, the summation factor of " -3 " is applied to the intermediate elements skew shown in Figure 12 B can be caused the right side Side intermediate elements 1208b is relative to the left side intermediate elements 1208a shifted left " 3 " individual grid cell so that in application drawing 12D After the shown summation factor, the centre between the center of the center of left side intermediate elements 1208a and the right intermediate elements 1208b Skew can be " 3 " individual grid cell.
Additionally, in Figure 12 B, between the center of the center of left side distally key element 1210a and the right distally key element 1210b The skew of distally key element can be " 3 " individual grid cell.Therefore, the summation factor of " -3 " is applied to remote shown in Figure 12 B Side key element skew can cause the right distally key element 1210b relative to the individual grid list in left side distally key element 1210a shifted lefts " 3 " The center of unit so that after the summation factor is applied, the center of left side distally key element 1210a and the right distally key element 1210b it Between distally skew can be " 0 " individual grid cell, as indicated in fig. 12d.
In the illustrated embodiment, the right nearside key element 1206b, the right intermediate elements being associated with right eye viewing frame 1208b and the right distally key element 1210b can respectively be moved to the left " 3 " individual grid cell, and be associated with left eye viewing frame Left side nearside key element 1206a, left side intermediate elements 1208a and left side distally key element 1210a can not in fig. 12d relative to Figure 12 B shift to obtain desired skew adjustment amount.In an alternative embodiment, in fig. 12d relative to Figure 12 B, nearside During skew, middle skew and distally skew can be by shifting left side nearside key element 1206a, the right nearside key element 1206b, the left side Between in key element 1208a, the right intermediate elements 1208b, left side distally key element 1210a and/or the right distally key element 1210b To adjust.
In addition, shown embodiment depicts the key element being associated with right eye viewing frame relative in left eye viewing frame Corresponding key element shifted left, focus is moved back to, and this can bring more key elements in the prospect perceived by beholder. However, as described above, in other embodiments, the key element being associated with right eye viewing frame can be watched relative to it in left eye Corresponding key element right shift in frame, to move forward focus, this can bring more in the background perceived by beholder Key element.
Therefore, some embodiments of the invention, three-dimensional video-frequency module (for example, three-dimensional video-frequency module 104) can quilt It is configured to pass to the left eye viewing frame with three-dimensional video-frequency and corresponding right eye viewing frame associated corresponding key element and/or pixel Between the offset applications unified summation factors adjust the focus being associated with three-dimensional video-frequency.
It will be appreciated by those skilled in the art that for this and other processes and method disclosed by the invention, can be with not With order come implement in process and method perform function.Additionally, it is summarized the step of and operation only property is carried as an example For, and some steps and operations can be optional, be combined into less step and operation or expand to additional step In operation, without departing from the essence of disclosed embodiment.
Figure 13 is the illustrative methods of the depth of the adjustment three-dimensional video-frequency of some embodiments according to present disclosure 1300 flow chart.In some embodiments, method 1300 can be by the three-dimensional video-frequency module of three-dimensional video-frequency module such as Fig. 1 104 implement.For example, three-dimensional video-frequency module 104 can be configured as computer instructions to perform as by method 1300 The operation of the depth for adjusting three-dimensional video-frequency represented by one or more frames.While shown as discrete block, but according to the phase The implementation of prestige, various frames can be divided into additional frame, be combined into less frame or be eliminated.
Method 1300 can start, and at frame 1302, can generate the left eye viewing frame of three-dimensional video-frequency.Left eye viewing frame can be with Including multiple left eyes viewing viewing frame key element.In some embodiments, left eye viewing frame key element can be included in left eye Essentially all of key element in viewing frame so that it is whole left that left eye viewing frame key element can substantially be covered with (encompass) Observe and see frame.In addition, left eye viewing frame key element can each include one or more pixels.
At frame 1304, the right eye viewing frame of three-dimensional video-frequency can be generated.Right eye viewing frame can correspond to left eye viewing Frame, and multiple right eye viewing frame key elements can be included.Each right eye viewing frame key element can correspond to left eye viewing frame key element In one.In some embodiments, right eye viewing frame key element can be included in the substantially institute in right eye viewing frame Some key elements so that right eye viewing frame key element can substantially be covered with whole right eye viewing frame.In addition, right eye viewing frame key element can Each to include one or more pixels.
At frame 1306, it may be determined that inclined between the corresponding right eye viewing frame key element of each left eye viewing frame key element Move.In some embodiments, skew can on a pixel-by-pixel basis be determined.At frame 1308, each skew can be answered With unified multiplication factor so that the depth that can be associated with three-dimensional video-frequency with substantially unified rescaling.
Therefore, method 1300 can be used for adjusting the depth of three-dimensional video-frequency.It will be appreciated by those skilled in the art that for this This and other processes and method of disclosure of the invention, can in a different order implement the work(performed in process and method Energy.Additionally, it is summarized the step of and operation only property is provided as an example, and some steps and operations can be optional, group In synthesizing less step and operation or expanding to additional step and operate, without departing from the sheet of disclosed embodiment Matter.
For example, in some embodiments, method 1300 can include and generation left eye viewing frame and right eye viewing frame phase The step of association.In some of these embodiments, can be according to one or more mode described above with reference to Fig. 1-11 Generation left eye viewing frame and right eye viewing frame.
Figure 14 is the illustrative methods of the focus of the adjustment three-dimensional video-frequency of some embodiments according to present disclosure 1400 flow chart.In some embodiments, method 1400 can be by the stereopsis of the three-dimensional video-frequency module 104 of such as Fig. 1 Frequency module is realized.For example, three-dimensional video-frequency module 104 can be configured as computer instructions to perform by method 1400 The operation of the focus for adjusting three-dimensional video-frequency that one or more frames are represented.While shown as discrete block, but according to expectation Implementation, various frames can be divided into additional frame, be combined into less frame or be eliminated.
Method 1400 can start, and at frame 1402, can generate the left eye viewing frame of three-dimensional video-frequency.Left eye watches frame Multiple left eye viewing frame key elements can be included.In some embodiments, left eye viewing frame key element can be included in left eye Essentially all of key element in viewing frame so that left eye viewing frame key element can substantially be covered with whole left eye viewing frame.Separately Outward, left eye viewing frame key element can each include one or more pixels.
At frame 1404, the right eye viewing frame of three-dimensional video-frequency can be generated.Right eye viewing frame can correspond to left eye viewing Frame, and multiple right eye viewing frame key elements can be included.Each right eye viewing frame key element can correspond to left eye viewing frame key element In one.In some embodiments, right eye viewing frame key element can be included in the substantially institute in right eye viewing frame Some key elements so that right eye viewing frame key element can substantially be covered with whole right eye viewing frame.In addition, right eye viewing frame key element can Each to include one or more pixels.
At frame 1406, it may be determined that inclined between the corresponding right eye viewing frame key element of each left eye viewing frame key element Move.In some embodiments, skew can on a pixel-by-pixel basis be determined.At frame 1408, can by unified summation because Son is applied to each skew.The unified summation factor can be applied so that each right eye viewing frame key element can be right relative to its The left eye viewing frame key element answered uniformly shifts substantially identical amount.Therefore, displacement can be adjusted and is associated with three-dimensional video-frequency Focus.
Therefore, method 1400 can be used for adjusting the focus of three-dimensional video-frequency.It will be appreciated by those skilled in the art that for this This and other processes and method of disclosure of the invention, can in a different order implement the work(performed in process and method Energy.Additionally, it is summarized the step of and operation only property is provided as an example, and some steps and operations can be optional, group In synthesizing less step and operation or expanding to additional step and operate, without departing from the sheet of disclosed embodiment Matter.
For example, in some embodiments, method 1400 can include and generation left eye viewing frame and right eye viewing frame phase The step of association.In some of these embodiments, can be according to one or more mode described above with reference to Fig. 1-11 Generation left eye viewing frame and right eye viewing frame.
Embodiment described herein can include using special or all-purpose computer, including various computer hardwares or soft Part module, as discussed in more detail below.
The embodiments described herein can be using for carrying or with the computer executable instructions being stored thereon Or the computer-readable medium of data structure is implemented.Such computer-readable medium can be can be by universal or special calculating Any usable medium that machine is accessed.As an example property and it is unrestricted, such computer-readable medium can include tangible meter Calculation machine readable storage medium storing program for executing, including RAM, ROM, EEPROM, CD-ROM or other optical disc storages, disk storage or other magnetic are deposited Storage equipment or any other storage medium, it can be used for carrying or storing computer executable instructions or data structure form Expectation program code, and can be accessed by universal or special computer.Combinations of the above can also be included in computer can Read in the range of medium.
Computer executable instructions can be performed by processing equipment, and can include for example cause all-purpose computer, specially A certain function or one group of instruction and data of function are performed with computer or dedicated treatment facility.Although with to architectural feature And/or method acts special language and describes theme, but it is to be understood that the theme defined in appended claims is not necessarily It is limited to above-mentioned specific features or action.Conversely, above-mentioned specific features and action are disclosed as implementing the exemplary shape of claim Formula
As it is used herein, term " module " or " component " can refer to the software object or journey for performing on a computing system Sequence.Different components described herein, module, engine and service may be implemented as object or the process for performing on a computing system (for example, as single thread).Although systems and methods described herein can use software implementation, the reality in hardware Apply or the combination of software and hardware is also possible and expected.In this manual, " computational entity " can be as this paper's Any computing system of preceding definition, module or the combinations of modulation run on a computing system.
All exemplary and conditional language as herein described is intended to for teaching object, to help reader to understand the present invention The conception contributed with this area is promoted by inventor, and the example and bar of these specific narrations should be to be construed as being without limitation of Part.Although embodiment of the present invention is described in detail, but it is to be understood that do not departing from the spirit and scope of the present invention In the case of, various changes can be carried out to it, is replaced and is changed.

Claims (24)

1. a kind of method, methods described includes：
Determine the movement between the first frame of monoscopic video and the second frame of the monoscopic video；
Based on the mobile analysis camera effect；
Frame is changed based on the movement and camera effect generation first；
Frame is changed based on the movement and camera effect generation second；
Based on the camera effect and at least one of the first modification frame and the second modification frame generation stereopsis The left eye viewing frame of frequency；And
It is described vertical based on the camera effect and the generation of at least one of the first modification frame and the second modification frame The right eye viewing frame of volumetric video.
2. the method for claim 1, also includes：
The left-half of left-half generation the first modification frame based on second frame；
The right half part of right half part generation the first modification frame based on first frame；
The left-half of left-half generation the second modification frame based on first frame；And
The right half part of right half part generation the second modification frame based on second frame.
3. method as claimed in claim 2, also includes：
Determine that the camera effect is to reduce effect；
Described first modification frame is appointed as by the left eye viewing frame based on the diminution effect；And
Described second modification frame is appointed as by the right eye viewing frame based on the diminution effect.
4. method as claimed in claim 2, also includes：
Determine that the camera effect is amplification effect；
Described first modification frame is appointed as by the right eye viewing frame based on the amplification effect；And
Described second modification frame is appointed as by the left eye viewing frame based on the amplification effect.
5. method as claimed in claim 2, also includes：
Ignore and vertically move to generate the left side of the first modification frame between first frame and second frame Point；And
Ignore and vertically move to generate the right side of the second modification frame between first frame and second frame Point.
6. the method for claim 1, also including based between first frame and second frame it is unified towards with The movement at the center of the associated scene of first frame and second frame is come to determine the camera effect be to reduce effect.
7. the method for claim 1, also including based on the unification between first frame and second frame away from The movement at the center of the associated scene of first frame and the second frame determines that the camera effect is amplification effect.
8. the method for claim 1, it is characterised in that determine movement between first frame and second frame also Including：
The first element that analysis is associated with the Part I of first frame and the corresponding Part I of second frame It is mobile；
The second key element that analysis is associated with the Part II of first frame and the corresponding Part II of second frame It is mobile；
The three elements that analysis is associated with the Part III of first frame and the corresponding Part III of second frame It is mobile；And
The 4th key element that analysis is associated with the Part IV of first frame and the corresponding Part IV of second frame It is mobile.
9. the method for claim 1, also includes：
Based on the mobile deduction scene between first frame and second frame and first frame and the second frame phase The subsequent frame of association；And
Based at least one of the subsequent frame generation first modification frame inferred and the second modification frame.
10. method as claimed in claim 9, also includes：
The left-half of left-half generation the first modification frame based on the subsequent frame inferred；
The right half part of right half part generation the first modification frame based on first frame；
The left-half of left-half generation the second modification frame based on first frame；And
The right half part of right half part generation the second modification frame based on the subsequent frame inferred.
11. methods as claimed in claim 10, also include：
Determine that the camera effect is to reduce effect；
Described first modification frame is appointed as by the left eye viewing frame based on the diminution effect；And
Described second modification frame is appointed as by the right eye viewing frame based on the diminution effect.
12. methods as claimed in claim 10, also include：
Determine that the camera effect is amplification effect；
Described first modification frame is appointed as by the right eye viewing frame based on the amplification effect；And
Described second modification frame is appointed as by the left eye viewing frame based on the amplification effect.
A kind of 13. computer-readable recording mediums, including it is configured as the computer executable instructions for making system perform operation, institute Stating operation includes：
Determine the movement between the first frame of monoscopic video and the second frame of the monoscopic video；
Based on the mobile analysis camera effect；
Frame is changed based on the movement and camera effect generation first；
Frame is changed based on the movement and camera effect generation second；
Based on the camera effect and at least one of the first modification frame and the second modification frame generation stereopsis The left eye viewing frame of frequency；And
It is described vertical based on the camera effect and the generation of at least one of the first modification frame and the second modification frame The right eye viewing frame of volumetric video.
A kind of 14. methods for analyzing video, methods described includes：
Second frame of the first frame and the video based on the video determines the most fast mobile key element in the video, at a slow speed Mobile at least one of key element and Primary element；And
Based on the most fast mobile key element, described to move slowly at least one of key element and the Primary element described to determine At least one of foreground and background in video.
15. methods as claimed in claim 14, also including making the most fast mobile key element be associated with the prospect.
16. methods as claimed in claim 14, also include：
Determine the Primary element substantially in the center of the scene being associated with first frame and second frame；And
The prospect is set to be associated with the Primary element substantially in the center of the scene based on the Primary element.
17. methods as claimed in claim 14, also include：
Determine the center that the Primary element is deviated substantially from the scene being associated with first frame and second frame；With And
The center in the scene is deviated substantially from based on the Primary element to make the background related to the Primary element Connection.
18. methods as claimed in claim 14, also include：
Determine unification of all key elements of first frame and second frame between first frame and second frame Moving direction；And
At least one of the prospect and the background are determined based on the unified moving direction.
19. methods as claimed in claim 18, also include：
It is determined that the most fast mobile key element and it is described move slowly at key element between speed difference；
The speed difference and threshold value are compared；And
If the speed difference is less than the threshold value, the system based on key element between first frame and second frame One moving direction, makes all key elements being associated with first frame and second frame be associated with the background.
20. methods as claimed in claim 14, also include：
Determine that all key elements of first frame and second frame are different between first frame and second frame Moving direction；And
At least one of the prospect and the background are determined based on the different moving direction.
21. methods as claimed in claim 20, also include：
It is determined that the most fast mobile key element and it is described move slowly at key element between speed difference；
The speed difference and threshold value are compared；And
If the speed difference is less than the threshold value, the phase based on key element between first frame and second frame Different moving direction, makes all key elements being associated with first frame and second frame be associated with the background.
22. methods as claimed in claim 20, also include：
It is determined that the most fast mobile key element and it is described move slowly at key element between speed difference；
The speed difference and threshold value are compared；And
If the speed difference is less than the threshold value, the phase based on key element between first frame and second frame Different moving direction, makes all key elements being associated with first frame and second frame be associated with the prospect.
23. methods as claimed in claim 14, also include：
The mobile degree between the background and the prospect is determined based on first frame and second frame；And
Left eye viewing frame and right eye viewing frame based on the mobile degree generation three-dimensional video-frequency.
A kind of 24. computer-readable recording mediums, including be configured as making system to perform operation can to analyze the computer of video Execute instruction, the operation includes：
Second frame of the first frame and the video based on the video determines the most fast mobile key element in the video, at a slow speed Mobile at least one of key element and Primary element；And
Based on the most fast mobile key element, described to move slowly at least one of key element and the Primary element described to determine At least one of foreground and background in video.
CN201580043225.8A 2014-06-10 2015-06-10 Stereoscopic video zooming and foreground and background detection in a video Pending CN106716995A (en)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
US14/301,146 US9661309B2 (en) 2014-06-10 2014-06-10 Stereoscopic video zooming
US14/301,143 2014-06-10
US14/301,143 US9693042B2 (en) 2014-06-10 2014-06-10 Foreground and background detection in a video
US14/301,146 2014-06-10
PCT/US2015/035186 WO2015191768A1 (en) 2014-06-10 2015-06-10 Stereoscopic video zooming and foreground and background detection in a video

Publications (1)

Publication Number Publication Date
CN106716995A true CN106716995A (en) 2017-05-24

Family

ID=54834271

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201580043225.8A Pending CN106716995A (en) 2014-06-10 2015-06-10 Stereoscopic video zooming and foreground and background detection in a video

Country Status (5)

Country Link
EP (1) EP3155809A4 (en)
JP (1) JP6490107B2 (en)
KR (1) KR101907127B1 (en)
CN (1) CN106716995A (en)
WO (1) WO2015191768A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110620916A (en) * 2018-12-29 2019-12-27 北京时光荏苒科技有限公司 Method and apparatus for processing image

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5028996A (en) * 1989-06-26 1991-07-02 Matsushita Electric Industrial Co., Ltd. Picture coding method
US7161614B1 (en) * 1999-11-26 2007-01-09 Sanyo Electric Co., Ltd. Device and method for converting two-dimensional video to three-dimensional video
CN101002232A (en) * 2004-08-12 2007-07-18 古鲁洛吉克微系统公司 Processing of video image
CN101479765A (en) * 2006-06-23 2009-07-08 图象公司 Methods and systems for converting 2D motion pictures for stereoscopic 3D exhibition
US20120127267A1 (en) * 2010-11-23 2012-05-24 Qualcomm Incorporated Depth estimation based on global motion
US20140104381A1 (en) * 2009-02-05 2014-04-17 Bitanimate, Inc. Two-dimensional video to three-dimensional video conversion method and system

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000261828A (en) * 1999-03-04 2000-09-22 Toshiba Corp Stereoscopic video image generating method
JP2006042280A (en) * 2004-07-30 2006-02-09 Nihon Brain Ware Co Ltd Image processing apparatus
US9232228B2 (en) * 2004-08-12 2016-01-05 Gurulogic Microsystems Oy Processing of image
US8330801B2 (en) * 2006-12-22 2012-12-11 Qualcomm Incorporated Complexity-adaptive 2D-to-3D video sequence conversion
JP2011114863A (en) * 2009-11-23 2011-06-09 Samsung Electronics Co Ltd Method for providing 3d image, method for converting 3d image, gui providing method, 3d display apparatus using the same, and system for providing 3d image
US9171372B2 (en) * 2010-11-23 2015-10-27 Qualcomm Incorporated Depth estimation based on global motion
JP2012213583A (en) * 2011-03-31 2012-11-08 Fujifilm Corp Stereoscopic display apparatus

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5028996A (en) * 1989-06-26 1991-07-02 Matsushita Electric Industrial Co., Ltd. Picture coding method
US7161614B1 (en) * 1999-11-26 2007-01-09 Sanyo Electric Co., Ltd. Device and method for converting two-dimensional video to three-dimensional video
CN101002232A (en) * 2004-08-12 2007-07-18 古鲁洛吉克微系统公司 Processing of video image
CN101479765A (en) * 2006-06-23 2009-07-08 图象公司 Methods and systems for converting 2D motion pictures for stereoscopic 3D exhibition
US20140104381A1 (en) * 2009-02-05 2014-04-17 Bitanimate, Inc. Two-dimensional video to three-dimensional video conversion method and system
US20120127267A1 (en) * 2010-11-23 2012-05-24 Qualcomm Incorporated Depth estimation based on global motion

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110620916A (en) * 2018-12-29 2019-12-27 北京时光荏苒科技有限公司 Method and apparatus for processing image

Also Published As

Publication number Publication date
JP2017523671A (en) 2017-08-17
KR20170035885A (en) 2017-03-31
EP3155809A1 (en) 2017-04-19
JP6490107B2 (en) 2019-03-27
EP3155809A4 (en) 2018-01-03
KR101907127B1 (en) 2018-10-11
WO2015191768A1 (en) 2015-12-17

Similar Documents

Publication Publication Date Title
US8711204B2 (en) Stereoscopic editing for video production, post-production and display adaptation
US9445072B2 (en) Synthesizing views based on image domain warping
KR20180099703A (en) Configuration for rendering virtual reality with adaptive focal plane
CN106664396A (en) Stereoscopic video generation
US10554954B2 (en) Stereoscopic focus point adjustment
CN106716995A (en) Stereoscopic video zooming and foreground and background detection in a video
CN106688230A (en) Stereoscopic depth adjustment and focus point adjustment
KR101192121B1 (en) Method and apparatus for generating anaglyph image using binocular disparity and depth information
Benzeroual et al. 3D display size matters: Compensating for the perceptual effects of S3D display scaling
US9693042B2 (en) Foreground and background detection in a video
US9674500B2 (en) Stereoscopic depth adjustment
Schild et al. Integrating stereoscopic video in 3D games
Alazawi Holoscopic 3D image depth estimation and segmentation techniques
US9661309B2 (en) Stereoscopic video zooming
Shibuhisa et al. 71.3: Study on Reducing the Cardboard Effect for Natural Perception Using Adaptive Disparity Mapping
Zilly Method for the automated analysis, control and correction of stereoscopic distortions and parameters for 3D-TV applications: new image processing algorithms to improve the efficiency of stereo-and multi-camera 3D-TV productions
Park et al. A method of frame synchronization for stereoscopic 3D video
Zilly Method for the automated analysis, control and correction of stereoscopic distortions and parameters for 3D-TV applications
Kim et al. Analysis of Space Roundness with Converged Stereo Camera for Realistic 3D Content Production
GB2535726A (en) An image processing method and apparatus for determining depth within an image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170524