WO2001039512A1 - Device and method for converting two-dimensional video to three-dimensional video - Google Patents
Device and method for converting two-dimensional video to three-dimensional video Download PDFInfo
- Publication number
- WO2001039512A1 WO2001039512A1 PCT/JP2000/008315 JP0008315W WO0139512A1 WO 2001039512 A1 WO2001039512 A1 WO 2001039512A1 JP 0008315 W JP0008315 W JP 0008315W WO 0139512 A1 WO0139512 A1 WO 0139512A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- amount
- parallax
- region
- video signal
- phase
- Prior art date
Links
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/189—Recording image signals; Reproducing recorded image signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
- H04N2013/0081—Depth or disparity estimation from stereoscopic image signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
- H04N2013/0085—Motion estimation from stereoscopic image signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
- H04N2013/0092—Image segmentation from stereoscopic image signals
Definitions
- the present invention relates to an apparatus and a method for converting a 2D image into a 3D image.
- the motion vectors between each image such as the motion vector when moving from image 1 to image 2 and the motion vector for moving from image 2 to image 3, are set in the screen. Is extracted for each motion vector detection area. Next, the subject (bird) region and the background (mountain) region are determined from the extracted motion vector. Then, the reference image is determined as one of the right and left eye images so that the subject is in front of the background, and the image delayed by the number of fields corresponding to the size of the motion vector is determined as the other.
- the current image which is the reference image
- an image (delayed image) delayed by a predetermined number of fields according to the size of the motion vector obtained from image 3 and image 4 is image 2.
- the reference image (image 4) is presented as the left-eye image
- the delayed image (image 2) is presented as the right-eye image from the direction of the motion vector.
- this method is referred to as the MTD method.
- the two-dimensional image is divided into a plurality of regions, and image features such as color components, high-frequency components, and contrast are extracted for each divided region.
- grouping is performed for each region to which the same object belongs.
- the depth is estimated based on information such as the average contrast and the average high-frequency component of the region, and the amount of parallax is calculated.
- a left-eye image and a right-eye image are horizontally shifted in opposite directions for each of the divided regions to generate a three-dimensional image.
- the left-eye image and the right-eye image created in this way are displayed on the three-dimensional display means and displayed three-dimensionally. This method will be referred to as the CID method.
- the optimal (target) field delay for the video is the average value of the horizontal vector obj-xvec [pixel] of the subject area obtained by subject background discrimination. / field] and the horizontal vector of the background area bg—xvec [pixel I field], and is determined by the following equation (1).
- the vector takes a positive value for rightward movement.
- d — target Mdly_sisa / (obj_xvec-bg_xvec) [field] ⁇ ⁇ ⁇ (1)
- Mdly-sisa represents a parallax amount [pi xel] that determines the stereoscopic effect generated by the MTD method, and its value is set in advance via a user interface or the like.
- the delay amount has been described using the target delay amount as an example, but in actual control, the number of fields to be delayed and the delay direction are determined by the actual delay amount delay obtained by temporally smoothing the target delay amount.
- the subject position control is used to correct the ambiguity of the object presentation position with respect to the tube surface that occurs when the MTD method is performed. That is, in the MTD method, the appearance differs depending on whether the subject moves or the background moves as shown in FIG. Therefore, in the subject position control, when the subject moves, the presenting position of the right-eye image is shifted to the right, and the presenting position of the left-eye image is shifted to the left to move the entire screen to the back.
- the number of pixels from the subject to the tube surface is equal to the number of pixels from the tube surface to the background.
- the presenting position of the right-eye image is shifted to the left, and the presenting position of the left-eye image is shifted to the right, so that the entire screen is brought forward.
- the number of pixels in the background is equal to the number of pixels in the background.
- the right eye water 51 ⁇ phase amount t_jhr and left eye horizontal phase amount t_phl calculated by the subject position control are the subject phase amount obj_sisa and the background phase amount bg—
- sisa is expressed by the following equation (3), it is expressed by the following equation (4).
- obj_sisa obj_xvec * delay [pixel] (3)
- bg-one sisa bg-one xvec * delay [pixel]
- t _phr (obj_sisa + bg—sisa) / 2 [pixel]
- the CID method divides a screen into multiple regions, estimates the depth of each region from image information and composition obtained from each region, and shifts each pixel in the screen based on the estimated depth. This is a technique for generating binocular parallax.
- the present applicant has also developed a CID method which is a further improvement of the CID method already developed.
- FIG. 3 shows a control procedure of the improved CID method (not known).
- one screen is divided into a plurality of areas, and information on high frequency, luminance contrast, and color (B-Y, R-Y components) is obtained from each area (step 1). Then, an estimated depth value of each area estimated from the information and the composition is obtained (step 2). If the obtained estimated depth value is simply converted to a shift amount, the distortion is noticeable in the converted image, so that distortion suppression processing is performed (step 3). A distance scale conversion is performed on the depth estimation value after the distortion suppression processing (step 4).
- the distortion suppression processing will be described.
- the difference in the amount of phase between adjacent regions is determined by the distortion tolerance h-supp-lev [Pixel ] It is controlled to be below. That is, first, the estimated depth is obtained by assigning it between Mfront and rear. The difference between the phase amounts of adjacent regions is determined from the phase amounts of the respective regions. Then, the maximum value of this difference is defined as h-dvma Lpixel].
- h-dv-max force distortion is greater than ⁇ -enclosure h-sup-lev [pixel]
- the following equation (5) is satisfied. Make Mfront and Mrear smaller toward 0 [pixe 1].
- the distance scale conversion method will be described.
- the parallax W between the corresponding points of the right-eye image (R image) and the left-eye image (L image) and the fusion position (distance from the tube surface that is actually seen) Yp are There is a nonlinear relationship.
- each variable represents the following value.
- W Parallax amount of corresponding point between left-eye image and right-eye image on display surface [ ⁇ ]
- Yp Distance from tube surface to fusion position [bandon]
- the maximum protrusion amount Ymax 'from the screen and the maximum depth Ymin' from the screen are specified, it will be supported if the estimated depth d mark th (having a value of 0 to 100) is determined.
- the depth Yp can be obtained by a simple scale conversion expressed by the following equation (12).
- the parallax amount ⁇ corresponding to Yp is obtained by the above equation (11).
- the depth estimation value can be converted into pixel units in consideration of spatial distortion.
- W "[255] is the disparity amount corresponding to Ymax '
- W" [0] is the disparity amount corresponding to Ymin.
- lev 255 depth / 100... (14)
- Dlev is defined by the following equation (17) and represents the number of steps on the parallax conversion table corresponding to the screen surface.
- the values actually used are determined Ymax ', Ymin', and C so that the inclination (step width) of the depth parallax conversion table before and after the screen does not greatly differ.
- the above-described distortion suppression processing using the linear operation is effective for pixel scale conversion, but cannot be said to be an effective means for distance scale conversion.
- the depth Yp and the disparity amount W [pixel] are not equal. It is linear and has the property that even if the depth estimation is the same value, for example, "1", the parallax is greatly different before and after the tube face. This tendency is noticeable in large-screen displays.
- the polygonal distance scale which is an improved version of the complete distance scale, introduces the pop-out ratio C to alleviate this characteristic.
- the maximum value h-dv—max [pixel] of the phase difference between adjacent regions is completely within the distortion tolerance h-supp—lev [pixel] even on a polygonal distance scale where the pop-out ratio C can be controlled. It cannot be suppressed (the principle of distortion suppression on the pixel scale cannot be faithfully realized). In order to realize this principle of distortion suppression, it is necessary to perform distortion suppression processing after distance scale conversion.
- a person perceives a sense of distance in stereoscopic vision due to a difference (occlusion) in a blind spot of an image entering each of the right and left eyes caused by a difference in the position of the left and right eyes.
- the MTD method could cover, but could not successfully convert a motionless video or a video with complicated motion into a 3D video.
- the parallax amount of the left and right eye images can be freely changed by the CID method, it is impossible for the left and right eyes of a person to make the blind spot portion that becomes a shadow of the subject look different due to the parallax.
- the disparity of the MTD method and the disparity of the CID method are reflected in the converted video, but if the input image is a still image, there is no disparity of the MTD method but only the disparity of the CID method.
- This invention uses the MTD method and the CID method together to convert 2D images to 3D images. It is an object of the present invention to provide a method for converting a 2D video to a 3D video, which can avoid a large difference in a stereoscopic effect of a converted video depending on an input video. Further, the present invention provides a method of converting a two-dimensional video into a three-dimensional video, which can suppress distortion of a converted image when converting a depth estimation amount into a parallax amount using a distance scale conversion. With the goal.
- An apparatus for converting a two-dimensional video to a three-dimensional video includes a field memory for storing a two-dimensional video signal input from a video signal source for each field, and a motion base corresponding to a motion between fields of the input video signal.
- the reading means for reading the video signal delayed from the field memory by the delay amount obtained from the motion vector of the area, and the direction of the horizontal component of the motion vector of each area detected by the motion vector detecting means
- One of the input video signal and the video signal read from the field memory is used as the left-eye video signal, and the other video signal is used as the right-eye video signal.
- Switching means for outputting as an image signal; image characteristic amount extracting means for extracting an image characteristic amount from the input image signal for each region of the input image; image characteristic amount for each region extracted by the characteristic amount extracting means.
- a parallax amount calculating unit that calculates a depth amount for each region of the input video, and calculates a parallax amount for each region from the calculated depth amount for each region; a parallax for each region calculated by the parallax amount calculating unit.
- the parallax amount correcting means for correcting the amount according to the size of the motion vector for each area detected by the motion vector detecting means, and the parallax amount of each area corrected by the parallax amount correcting means, It is characterized by comprising phase control means for correcting the phase of each area of the right-eye video and the left-eye video output by the switching means and outputting the corrected three-dimensional video signal.
- each area calculated by the parallax amount calculating means is used.
- Means for calculating the difference parallax amount for each region by subtracting the parallax amount corresponding to the size of the motion vector of the corresponding region from the parallax amount for each region, and the maximum difference parallax amount between adjacent regions The one provided with means for changing the dynamic range so as to calculate the difference parallax amount for each region so that the value falls within a predetermined range is used.
- the surplus amount of parallax is calculated as It is preferable to provide a means for reducing the amount of delay by a corresponding amount.
- the first step is to store the video signal in the field memory for each field.
- the second step is to detect the motion vector corresponding to the movement between the fields of the input video signal for each area of the input video. 3rd step of reading out from the field memory a video signal that is delayed from the input video signal by the amount of delay obtained from the motion vector of each area detected in the 2nd step from the input video signal. The difference between the input video signal and the video signal read from the field memory is determined based on the direction of the horizontal component of the motion vector of each area detected in the second step.
- the seventh step for example, for each area calculated in the sixth step Calculating the difference amount of parallax for each region by subtracting the amount of parallax corresponding to the magnitude of the motion vector of the corresponding region from the amount of parallax of the corresponding region, and determining the maximum value of the difference of the amount of difference parallax between adjacent regions.
- a method including a step of calculating a difference parallax amount for each region by changing a dynamic range so as to fall within a range is used.
- the method includes a step of reducing the amount of delay by an appropriate amount.
- the first feature is to extract image feature amounts related to perspective of a video for each of a plurality of parallax calculation regions set in the screen, and to generate a depth estimation value for each parallax calculation region based on the extracted image feature amounts. Step, performing a distance scale conversion using a dynamic range defined by a predetermined maximum pop-out amount and a predetermined maximum depth amount on each of the estimated depth values, thereby obtaining a tentative target phase for each parallax calculation region.
- a second step for obtaining the amount a third step for obtaining the maximum value of the phase difference between adjacent parallax calculation areas based on the provisional target phase amount for each parallax calculation area, phase A force in which the maximum value of the difference is within a predetermined allowable distortion range, a fourth step of determining whether or not the force is different, and the maximum value of the phase difference between adjacent parallax calculation regions is out of the predetermined allowable distortion range.
- a dynamic range in which the phase difference between the parallax calculation regions is within the distortion allowable range is searched, and a distance scale conversion is performed on each of the estimated depth values using the searched dynamic range, After a temporary target phase amount is obtained for each parallax calculation area, a fifth step is performed to shift to a third step.
- the distance scale conversion is a method of converting the estimated depth value into a pixel unit (amount of parallax) in consideration of the fusion position.
- a method of linearly converting the estimated depth value into pixel units (parallax) is called pixel scale conversion.
- FIG. 1 is a schematic diagram for explaining a conventional MTD method.
- FIG. 2 is a schematic diagram for explaining subject position control.
- FIG. 3 is a flowchart showing a control processing procedure of the conventional CID method.
- FIG. 4 is a schematic diagram for explaining dynamic range suppression processing in the conventional CID method.
- FIG. 5 is a graph showing the relationship between the amount of parallax W and its fusion position Yp.
- FIG. 6 is a graph for explaining the complete distance scale conversion.
- FIG. 7 is a graph for explaining the polygonal distance scale conversion.
- Fig. 8 is a graph showing the characteristic that the depth Yp and the amount of parallax W [pixel] are non-linear, so that even if the estimated depth is the same value, for example, "1", the amount of parallax greatly differs before and after the tube surface. It is.
- FIG. 9 is a flowchart showing a control procedure of the CID method according to the first embodiment of the present invention.
- FIG. 10 is a flowchart showing details of the distance scale conversion and distortion suppression processing in step 13 of FIG.
- Fig. 11 is a graph showing that the depth relationship between the front and back of the tube surface is maintained even when the dynamic range changes by introducing the method of maintaining the distance ratio.
- FIG. 12 is a schematic diagram showing a case where the polygonal line distance scale conversion only for maintaining the ratio of the ratio is performed and a case where the process for maintaining the ratio of the distance is further performed.
- FIG. 13 is a diagram illustrating a schematic configuration of a 2DZ3D conversion device according to the second embodiment.
- FIG. 14 is a flowchart showing the operation of the 2DZ3D conversion device of FIG.
- FIG. 15 is a diagram illustrating a schematic configuration of a 2DZ3D conversion device according to the second embodiment.
- FIG. 16 is a flowchart showing the overall control processing procedure of the integrated phase control.
- FIG. 17 is a schematic diagram showing the overall behavior of the integrated phase control.
- FIG. 18 is a schematic diagram showing the behavior of each area during the integrated phase control.
- FIG. 19 is a flowchart showing a detailed processing procedure of steps 53, 54, and 55 in FIG.
- FIG. 20 is a schematic diagram showing an example of distortion suppression performed in integrated phase control.
- FIG. 21 is a schematic diagram showing a case where the phase adjustment by the MTD method is necessary and a case where it is not necessary.
- FIG. 22 is a schematic diagram for explaining the phase suppression processing of the MTD method.
- FIG. 23 is a flowchart illustrating a delay amount suppression processing procedure.
- FIG. 9 shows a control procedure of the CID method according to the first embodiment of the present invention.
- one screen is divided into a plurality of areas, and information on high frequency, luminance contrast, and color (B-Y, R-Y components) is obtained from each area (step 11).
- an estimated depth value of each area estimated from the information and the composition is obtained (step 12).
- a target phase amount is obtained (step 13).
- FIG. 10 shows details of the distance scale conversion and distortion suppression processing in step 13 of FIG.
- distance scale conversion is performed in the dynamic range specified by Mfront and Mrear to obtain a tentative target phase amount (steps 21 and 22). Provisional eyes obtained The maximum value h-dv-max [pixel] of the phase difference between adjacent regions is calculated based on the target phase amount (step 23).
- step 24 It is determined whether or not the maximum value h-dv-max [pixel] of the phase difference between the adjacent regions is within the distortion allowable range h-supp-lev [pixel] (step 24). If it is within the allowable range, the provisional target phase amount is set as the true target phase (step 27).
- step 25 the processing of step 25 is referred to as a sequential search processing. Details of the sequential search processing will be described later.
- step 26 After changing front and rear so that the distance ratio between front and rear determined by the sequential search processing becomes the distance ratio specified by the user (step 26), return to step 22 and perform further distance scale conversion. .
- Step 2 Perform the processing of 2, 2, 3, 24, 25, and 26 so that the maximum value of the phase difference between adjacent regions, h—dv—max [pixel], is within the distortion tolerance h—supp—lev [pixel]. Repeat until the final target phase amount is obtained. Whenever the dynamic range is changed in this way, the distance scale conversion is performed so that the observer perceives a stereoscopic image according to the estimated depth without being influenced by the spatial distortion of the stereoscopic display. This is in order to accurately realize Kale's principle.
- the range is determined by a sequential search process in order to increase the dynamic range defined by the front and rear values.
- the amount of parallax can be calculated using the depth parallax conversion equation (Equation 15), but it is more efficient to use the previously calculated parallax conversion table W ⁇ as shown below. is there.
- phase W "(lev) ⁇ ⁇ ⁇ (19)
- the optimal method of searching for the front value and the rear value depends on the relationship between the phase amounts of the two regions where the phase difference between adjacent regions is the largest. There are three types below.
- Max — lev and Min — lev are set so as to maintain the distance ratio specified by the user in the sequential search process, that is, to maintain the relationship of the following equation (2 0). Get closer to Dlev.
- the above equation (20) is synonymous with the processing of step 26 in FIG. In the first case and the second case, in order to reduce the amount of computation, the distance ratio change processing is not performed during the sequential search processing, but is performed in step 26 in FIG.
- the reason for maintaining this distance ratio on the distance scale is to maintain the depth relationship before and after the tube surface even when the dynamic range changes. Specifically, as shown in Fig. 11, if the distance before the screen is reduced by 20% in the first case, the distance after the screen is also reduced by 20%, Maintain the relationship before and after. If this distance ratio is maintained on the parallax amount conversion table, the relationship of the pop-out amount ratio C can also be maintained. As a result, a viewer who tends to perceive the space based on the relative depth can be presented with a converted image that does not cause any discomfort.
- the distance ratio is not maintained in the first and second cases, but only the ratio ratio is maintained.
- FIG. 12 shows a case in which the polygonal distance scale conversion is performed only for maintaining the ratio, and a case in which processing for maintaining the distance ratio is further performed.
- the depth estimation value and the parallax conversion table are correlated by separate range conversion using the screen depth estimation value as a boundary.
- the parallax amount conversion table of the estimated depth value can be handled by one range conversion.
- the figure The function lev (phase) in 1 2 represents the inverse function of the above equation (1 9), and means that the number of stages in the disparity amount conversion table is obtained from the phase amount phase [pixel].
- 1 is a video supply source serving as a video signal supply means for a VTR, CD-ROM, TV broadcast, etc.
- 2 is a two-dimensional video signal supplied from the video source 1, that is, a three-dimensional video signal.
- 2D to 3D conversion device that converts to left-eye video signal L and right-eye video signal R
- 3 is a 2D / 3D converter 2
- 2 is an image splitter system that displays 3D video signals output from 2 This is a three-dimensional display means using.
- Reference numeral 4 denotes a field memory for storing a video signal from the video source 1 in units of one finolade unit
- 5 denotes a motion vector detecting means for detecting a motion vector from the video signal from the video source 1.
- the color extracting means 6, the contrast extracting means 7, and the high-frequency component extracting means 8 constitute an image feature extracting means.
- Reference numeral 9 denotes a motion vector calculation means for obtaining the direction of motion and the motion amount (magnitude of the motion vector) from the motion vector detected by the motion vector detection means 5 and outputting the motion vector; and 10 denotes a current image.
- Memory control means for reading the delayed image (delayed image) from the field memory 4 by the number of fields corresponding to the motion amount output from the motion amount calculation means 9 with reference to the motion amount calculation means 9.
- Switching means for switching which of the reference image (current image) and the delayed image is output as the left-eye video signal L and which is output as the right-eye video signal R based on the direction of the output motion. It is.
- 14 is a second depth map creating means for calculating depth information from the contrast information extracted by the contrast extracting means 7 for each area grouped by the grouping means 12 and creating a depth map
- 15 is a grouping means 1
- Third depth map creating means for calculating depth information from the high frequency component information extracted by the high frequency component extracting means 8 for each area grouped in 2 and creating a depth map
- 16 is predetermined composition information and group
- the fourth depth map creating means calculates depth information from the area information grouped by the grouping means 12 and creates a depth map.
- the combined map created by the combined map creation unit 17, the disparity amount calculation unit that calculates the amount of parallax for each of the predetermined disparity amount calculation areas, and 19 is each disparity amount calculated by the disparity amount calculation unit 18
- This is a water-standing setting means for shifting the left and right eye images output from the switching means 11 on a pixel-by-pixel basis in the horizontal direction, based on the amount of parallax for each area, and combining them.
- parallax amount calculation unit 18 outputs to the horizontal position setting unit 19 according to the amount of motion in order to consider the amount of frame delay performed based on the amount of motion calculated by the amount of motion calculation unit 9.
- the amount of parallax has been corrected, that is, reduced.
- FIG. 14 shows the operation of the 2D / 3D conversion device 2.
- the video signal of the video source 1 is stored in the field memory 4 on a field-by-field basis (step 31).
- the motion vector is detected from the two-dimensional video signal of the video source 1 by the motion vector detection means 5, and the motion amount and the direction of the motion vector are calculated by the motion amount calculation means 9 (step 3). 2).
- the motion vector detection means 5 compares the current image with the image one field before, and The movement amount and the direction of the subject in the image are extracted as a motion vector.
- the switching means 11 outputs one of the reference image and the delayed image as a left-eye video signal L and the other as a right-eye video signal based on the direction of motion of the motion vector detected in step 3 2. Output as R.
- steps 31 to 33 correspond to the operations of the MTD method.
- an image feature is extracted based on the two-dimensional video signal from the video source 1 (step 34).
- the color extracting means 6 extracts color information for each image feature amount detection area.
- the contrast extracting means 7 extracts a contrast for each image feature amount detection area.
- the high frequency extraction unit 8 extracts a high frequency component for each image feature amount detection area.
- the grouping means 12 includes color information for each image feature amount detection area extracted by the color extraction means 6 and motion information detected in step 32 for use in discriminating a subject or a background. Group amounts and regions in the image.
- a depth map is created (step 35). That is, the first depth map creating means 13 generates the first depth map based on the motion amount of the motion vector calculated by the motion amount calculating means 9 and the group data obtained by the grouping means 12. create.
- the second depth map creating means 14 is configured to generate a second depth map based on the contrast for each image feature amount detection area extracted by the contrast extracting means 7 and the grouping information obtained by the grouping means 12. Create a map.
- the third depth map creating means 15 is configured to generate a third depth map based on the high frequency components for each image feature amount detection area extracted by the high frequency extracting means 8 and the grouping information obtained by the grouping means 12. Create a map.
- the fourth depth map creating means 16 uses a predetermined screen composition (for example, if a landscape image is mainly used, the lower part of the screen is the ground, the upper part of the screen is empty, and the subject is the object in the center of the screen). And a fourth depth map is created based on the group information obtained by the grouping means 12.
- a synthetic depth map is created (step 36).
- the composite map creating means 17 weights and adds the first to fourth depth maps created by the first to fourth depth map creating means 13, 14, 15, and 16, Create a composite depth map.
- the parallax amount calculating means 18 calculates the parallax amount between the left-eye image and the right-eye image for each predetermined parallax amount calculation area based on the synthetic depth map created by the synthetic map creating means 17. calculate. Steps 34 and 37 described above correspond to the operation of the CID method.
- the parallax calculating means 18 corrects the parallax for each parallax calculating area calculated in step 37 according to the motion vector of the motion vector calculated by the motion calculating means 9. Specifically, each parallax amount calculated in step 37 is reduced by the parallax amount corresponding to the delay amount of the delayed image with respect to the reference image.
- the left-eye image L and the right-eye image R are shifted horizontally according to the corrected amount of parallax (step 39).
- the horizontal position setting unit 19 shifts the left-eye image L and the right-eye image R output from the switching unit 11 based on the amount of parallax corrected in step 38, for example, by horizontal shift for each pixel. I do.
- Figure 15 shows a device that converts 2D video to 3D video (2DZ 3D conversion device). 1 shows the configuration.
- reference numeral 101 denotes a video supply source serving as a video signal supply means for VTR, CD-ROM, TV broadcast, etc.
- reference numeral 102 denotes a two-dimensional video signal supplied from the video source 1 for three-dimensional video.
- 2D / 3D converter for converting signals, that is, left-eye video signal L and right-eye video signal R, 103 is an image splitter that displays a 3D video signal output from 2DZ 3D converter 2 This is a three-dimensional display means using a method or the like.
- the configuration of the conversion device 102 will be described.
- 104 is a field memory that stores the video signal from the video source 101 in units of one field
- 105 is a motion vector that detects a motion vector from the video signal from the video source 101. It is a detecting means.
- Reference numeral 106 denotes an image feature amount extraction unit that extracts image feature amounts such as color components, contrast, and high-frequency components from the video supplied from the video supply source 101 for each region.
- Reference numeral 110 denotes a delay amount calculating unit that calculates a delay amount from the motion vector detected by the motion vector detecting unit 105.
- Reference numeral 107 denotes a memory control means for reading, from the field memory 104, an image (delayed image) delayed by the number of fields corresponding to the delay amount calculated by the delay amount calculating means 110 based on the current input image. It is.
- Reference numeral 108 denotes a left-eye video signal L and any one of a right-eye video signal R of the input image and the delayed image based on the direction of motion output from the delay amount calculating means 110. This is switching means for switching whether or not to perform.
- Reference numeral 109 denotes a depth estimation value calculation unit that calculates a depth estimation value for each region based on the image feature amount for each region extracted by the image feature extraction unit 106.
- 1 1 1 calculates the parallax amount (phase amount) by the CID method for each area based on the depth estimation value for each area calculated by the depth estimation value calculation means, and outputs the parallax amount (phase amount) from the delay amount calculation means 110.
- each of the regions (for example, pixels) of the image for the left and right eyes output from the switching means 108 is horizontally shifted. This is a stereoscopic video compositing means that allows the user to fit the footage.
- the 112 is a disparity amount for controlling a delay amount based on the integrated disparity amount calculated by the disparity amount calculation unit 111 and the MTD method disparity amount output from the delay amount calculation unit 110. It is a monitoring means.
- a three-dimensional space is reproduced based on a depth estimation value obtained by the CID method.
- a stereoscopic video is presented by adding occlusion by the MTD method to the CID method.
- the phase of the MTD method (the amount of parallax: the phase generated as a result of the field delay) is subtracted from the phase (the amount of parallax) of each region calculated by the CID method.
- the phase of each region should be the same as the phase by the CID method even after using both the CID method and the CID method. For this reason, the amount of phase generated by the MTD method and the CID method is controlled by the following priorities.
- Priority 1 Maximum range of phase amount set by user Urange [pixel]
- Priority 2 Restriction of image distortion due to phase shift in converted image h_supp one lev [pixe ⁇ ]
- Priority 4 Phase amount generated by MTD method that does not exceed Urange dly—s isa [pixel]
- Priority 1 the highest priority, ensures that the combined phase does not exceed the maximum range Urange of the phase amount set by the user.
- Priority 2 ensures that image distortions produced by integrated phase control (especially the CID method) are within certain limits (h-supp-lev).
- Priority 3 means that the depth estimate (depth shape) for each region calculated by the CID method is preserved even after using both the MTD method and the CID method.
- Priority 4 is that the amount of disparity generated by the MTD method does not exceed Urange Guarantee.
- Priority 5 which has the lowest priority, means that the phase generated by the CID method has a different value from the phase generated by the CID method alone when used in combination with the MTD method.
- FIG. 16 shows an integrated phase control processing procedure that maintains the depth shape of the CID method.
- FIG. 17 shows the phase behavior at the time of implementation.
- an image feature amount is extracted by the image feature extraction means 106 (step 51).
- the estimated depth value calculating means 109 estimates the depth of the CID method based on the image feature amount extracted by the image feature extracting means 106 (step 52). That is, the calculated frequency, contrast, composition weight, and subject background discrimination result weight are added at an appropriate ratio to obtain an estimated depth value.
- the addition ratio is made variable according to the speed of movement in the video. Specifically, in order to compensate for the tendency of the value of the high frequency component falling due to fast movement, the rate of addition of the high frequency component is reduced as the movement becomes faster.
- the depth estimation value obtained in this way is subjected to distance scale conversion (complete distance scale conversion or polygonal distance scale conversion) in Ufront and Urear, and the phase amount of each region is obtained by the CID method (step 53).
- Distortion suppression processing is performed so that the difference between the quantities is less than h—supp—lev [pixel] (Step 55) 0
- the phase after the distortion suppression processing in FIG. 17 indicates that the distortion suppression processing is performed on the differential phase.
- the maximum projecting phase ufront [pixel] and the maximum depth phase urear [pixel] of the phase (integrated phase) obtained by integrating the MTD method and the CID method after the above-described distortion suppression processing are represented by the loop shown in Fig. 19 Required by processing.
- FIG. 19 shows details of the processing of steps 53, 54, and 55 in FIG. These processes are performed by the parallax amount calculation means 111.
- Ufront and Urear set by the user are set in the variables Ufront 'and Urear' (step 61), and then distance scale conversion is performed using the dynamic range defined by Ufront 'and Urear' to determine the CID phase amount.
- Get step 62.
- a temporary difference phase is obtained by subtracting the MTD phase from the CID phase (step 63).
- the maximum value of the phase difference between adjacent regions obtained from the provisional difference phase amount, h—dv-max [pixe 1], (the maximum value of the difference in the amount of difference phase between adjacent regions) is obtained (step 6 4). Then, go to Step 65.
- step 65 when such a loop processing is performed, the maximum value of the phase difference calculated in the previous step 64 h—dv—max [pixel] force The phase difference calculated in the current step 64 J is smaller than the maximum value of h-dv-max [pixel].
- step 65 When the above loop processing is not performed, the result of step 65 is NO, and the maximum value of the phase difference calculated in step 64 of this time h—dv—max [pixel] I distortion tolerance h—suppp—lev It is determined whether it is within [pixel] (step 66). If it is within the range, the provisional difference phase is set as the true target phase (step 72). Conversely, if it is out of the range, it is determined whether or not the number of loops is within the limit number of loops in order to reduce the CPU load (step 67). If the number of loops is larger than the limit number of loops, a true target phase is obtained by forced distortion suppression processing for the provisional difference phase, which will be described later (step 73).
- the temporary difference phase is saved (step 68), and Ufront 'and Urear use the phase difference between adjacent regions until h-supp-lev or less.
- the optimum ufront and urear values are obtained by gradually reducing the specified dynamic range (step 69). This process is hereinafter referred to as a sequential search process. Details of the sequential search processing will be described later.
- the distance ratio between ufront and urear obtained in the sequential search process is changed so as to be the distance ratio specified by the user (step 70).
- the dynamic range is changed by setting the obtained ufront and urear to Ufront 'and Urear' (step 71), and the process returns to step 62 to perform distance scale conversion again.
- Step 6 Steps 2 to 7 Repeat the series of steps 1 to 1 until the maximum value of the phase difference h — dv_max [pixel] within the allowable range of distortion h —supp—lev [pixel] is reached. The final target phase amount is obtained.
- the first interruption occurs when the number of loops reaches the limited number of loops in step 67 to reduce the CPU load.
- the temporary difference phase is subjected to pixel-scale distortion suppression processing as shown in the following equation (21), which is synonymous with equation (6), and the ufront and urear values are determined.
- df ufront and df urear are the maximum and minimum values of the temporary difference phase, respectively. And is obtained at the stage of calculating the maximum phase difference between adjacent regions with respect to the difference phase. Then, the differential phase falls within the range newly obtained in this way. It should be noted that there is no problem even if the processing of maintaining the distance ratio represented by the following equation (2 2) is performed on ufront and urear in the above equation (2 1).
- Dlev flev (ufront)-Dlev ⁇ : fDlev-lev (urear) ⁇ ⁇ ⁇ -(22)
- the second interruption is that in step 65, when the loop processing of step 62 to step 71 is performed, the maximum value of the phase difference calculated in the previous step 64 h 1 d V—max [pixel] Force Occurs when the phase difference calculated in step 64 of this time is smaller than the maximum value h—dv—max [pixel].
- This interruption occurs when the phase difference between adjacent regions in the current loop, h_dv__max, is not smaller than the value obtained in the previous loop, even though the dynamic range is sufficiently reduced. It is. This occurs because the phase generated by the MTD method is not changed by the distortion suppression processing. That is, as shown in Fig. 20, when the phase difference between the subject and the background generated by the MTD method is large, even if the number of loops is increased, the phase difference of the MTD phase hinders the dynamic range. Consequently, the difference phase will not be less than h-supp-lev. In such a case, the processing is interrupted, and the dynamic range is changed by the same processing as in step 73 (step 74).
- the dynamic range is changed for the temporary save difference phase saved in step 68.
- the dynamic range is changed with respect to the temporary save difference phase because the phase of the MTD method affects the shape of the difference phase and the dynamic range of the difference phase each time the distortion suppression loop is repeated. This is to reduce the tendency to decrease.
- the disparity amount (two-field delay amount X horizontal motion vector value of the region) that the region originally has is smoothed between adjacent regions as the MTD phase ph—ratdj of each region.
- the latter value is used.
- the phases of the MTD method and the CID method must be similar. For this reason, as shown in Fig. 16, in the integrated phase, depth estimation is performed in consideration of the result of subject background discrimination so that the CID phase increases in the region where the MTD phase increases.
- the search direction is determined according to the values of the CID phase in the two regions where the phase difference determined in (1) is maximum.
- phase difference h-dv-max obtained in 76 is determined in the following order.
- the object position control is not used in the MTD method.
- the phase generated by the MTD method is determined in advance by the user. It may exceed the specified maximum projecting phase amount Ufront Cpixel] and the maximum depth phase amount Urear [pixel].
- Figure 21 shows the phase behavior when such a phenomenon occurs.
- the OK mark at the right end in the figure indicates that the integrated phase, which is the sum of the MTD phase and the difference phase, is within the dynamic range Urange predetermined by the user, and the NG mark indicates that the integrated phase indicates the dynamic range Urange. Indicates exceeding.
- Urear is about the same as the distance between the eyes, the distance scale cannot define the depth beyond the distance between the eyes. If the NG phenomenon is maintained even after the distortion suppression processing, the principle of stereoscopic reproduction in Urange, which is a major premise of integrated phase, is not maintained.
- the parallax amount Mdly-sisa which determines the stereoscopic effect generated by the MTD method, to be small beforehand so that the NG phenomenon does not occur. It is hard to say that it is preferable because the three-dimensional effect by the MTD method is lost. Therefore, in order to compensate for the large Mdly-sisa, the occurrence of the NG phenomenon is recognized to some extent, and control is required to reduce the target delay dly-target only when a phase exceeding Ufront and Urear occurs (Fig. 22)).
- FIG. 23 shows a control processing procedure for realizing the processing of FIG. 22 (a control processing procedure performed by the parallax amount monitoring means 112).
- the target delay is reduced when the integrated phase of each region (sum of the actual phase and the phase based on the actual delay) exceeds Ufront and Urear.
- step 81 it is necessary to calculate the phase amount by the MTD method for each field subject region and background region (step 81).
- To calculate the phase amount of the current field In order to improve the accuracy, we use the actual phase phase [pixel] and the actual delay amount delay [field].
- the actual parallax obj_s isa '[pixe] of the subject area and the actual parallax bg—sisa' [pixel] of the background area caused by the field delay of the MTD method are unknown NG.
- the actual parallax ng—si sa '[piexel] of the region is calculated by the following equation (23).
- obj— sisa ' obj— vect * delay
- nasej ⁇ (u _j? hasej-Urear) for Urear> u _phasej
- the over-max p obtained by the above equation (25) is subtracted from the absolute value of the actual field disparity dly — sisa '[pixel] generated by the field delay, and the MTD method is used.
- the magnitudes of dly-target 'and the target delay amount before suppression dly-target are compared.
- phase of the MTD method is suppressed based on the actual delay amount and the actual parallax. It is possible.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Processing Or Creating Images (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP00977933A EP1235438B1 (en) | 1999-11-26 | 2000-11-24 | Method for converting two-dimensional video to three-dimensional video |
US10/130,427 US7161614B1 (en) | 1999-11-26 | 2000-11-24 | Device and method for converting two-dimensional video to three-dimensional video |
DE60014420T DE60014420T2 (de) | 1999-11-26 | 2000-11-24 | Verfahren zur 2d/3d videoumwandlung |
AT00977933T ATE278298T1 (de) | 1999-11-26 | 2000-11-24 | Verfahren zur 2d/3d videoumwandlung |
Applications Claiming Priority (8)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP33682499 | 1999-11-26 | ||
JP11/336824 | 1999-11-26 | ||
JP2000-58047 | 2000-03-02 | ||
JP2000058047 | 2000-03-02 | ||
JP2000109211A JP2001320731A (ja) | 1999-11-26 | 2000-04-11 | 2次元映像を3次元映像に変換する装置及びその方法 |
JP2000-109210 | 2000-04-11 | ||
JP2000-109211 | 2000-04-11 | ||
JP2000109210A JP4036599B2 (ja) | 2000-04-11 | 2000-04-11 | 2次元映像を3次元映像に変換する方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2001039512A1 true WO2001039512A1 (en) | 2001-05-31 |
Family
ID=27480553
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2000/008315 WO2001039512A1 (en) | 1999-11-26 | 2000-11-24 | Device and method for converting two-dimensional video to three-dimensional video |
Country Status (6)
Country | Link |
---|---|
US (1) | US7161614B1 (ja) |
EP (1) | EP1235438B1 (ja) |
AT (1) | ATE278298T1 (ja) |
DE (1) | DE60014420T2 (ja) |
TW (1) | TW521519B (ja) |
WO (1) | WO2001039512A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1489857A1 (en) * | 2002-03-27 | 2004-12-22 | Sanyo Electric Co., Ltd. | 3-dimensional image processing method and device |
CN101815225A (zh) * | 2009-02-25 | 2010-08-25 | 三星电子株式会社 | 生成深度图的方法及其设备 |
US8131064B2 (en) | 2002-03-27 | 2012-03-06 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
CN102427550A (zh) * | 2011-12-09 | 2012-04-25 | 彩虹集团公司 | 一种实现2d转3d的方法 |
Families Citing this family (120)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003018619A (ja) * | 2001-07-03 | 2003-01-17 | Olympus Optical Co Ltd | 立体映像評価装置およびそれを用いた表示装置 |
EP1370075B1 (en) * | 2002-06-06 | 2012-10-03 | Accenture Global Services Limited | Dynamic replacement of the face of an actor in a video movie |
CN1675937B (zh) | 2002-08-20 | 2011-08-24 | 江良一成 | 生成立体图像的方法和装置 |
US20060103664A1 (en) * | 2002-08-27 | 2006-05-18 | Sharp Kabushiki Kaisha | Contents reproduction device capable of reproducing a contents in optimal reproduction mode |
EP1587329B1 (en) * | 2003-01-20 | 2015-04-15 | Sanyo Electric Co., Ltd. | Three-dimensional video providing method and three-dimensional video display device |
JP3700707B2 (ja) * | 2003-03-13 | 2005-09-28 | コニカミノルタホールディングス株式会社 | 計測システム |
KR100505334B1 (ko) * | 2003-03-28 | 2005-08-04 | (주)플렛디스 | 운동 시차를 이용한 입체 영상 변환 장치 |
AU2004237705B2 (en) * | 2003-05-02 | 2009-09-03 | Yoostar Entertainment Group, Inc. | Interactive system and method for video compositing |
US8027531B2 (en) * | 2004-07-21 | 2011-09-27 | The Board Of Trustees Of The Leland Stanford Junior University | Apparatus and method for capturing a scene using staggered triggering of dense camera arrays |
KR100649523B1 (ko) * | 2005-06-30 | 2006-11-27 | 삼성에스디아이 주식회사 | 입체 영상 표시 장치 |
US8330801B2 (en) * | 2006-12-22 | 2012-12-11 | Qualcomm Incorporated | Complexity-adaptive 2D-to-3D video sequence conversion |
EP2289235A4 (en) | 2008-05-20 | 2011-12-28 | Pelican Imaging Corp | RECORDING AND PROCESSING IMAGES BY MONOLITHIC CAMERA ARRANGEMENT WITH HETEROGENIC IMAGE TRANSFORMER |
US11792538B2 (en) | 2008-05-20 | 2023-10-17 | Adeia Imaging Llc | Capturing and processing of images including occlusions focused on an image sensor by a lens stack array |
US8866920B2 (en) | 2008-05-20 | 2014-10-21 | Pelican Imaging Corporation | Capturing and processing of images using monolithic camera array with heterogeneous imagers |
JP5083052B2 (ja) * | 2008-06-06 | 2012-11-28 | ソニー株式会社 | 立体視画像生成装置、立体視画像生成方法およびプログラム |
EP2292017A1 (en) * | 2008-06-13 | 2011-03-09 | Thomson Licensing | System and method for marking a stereoscopic film |
WO2010024479A1 (en) * | 2008-08-26 | 2010-03-04 | Enhanced Chip Technology Inc. | Apparatus and method for converting 2d image signals into 3d image signals |
WO2010025458A1 (en) * | 2008-08-31 | 2010-03-04 | Mitsubishi Digital Electronics America, Inc. | Transforming 3d video content to match viewer position |
CN101673395B (zh) * | 2008-09-10 | 2012-09-05 | 华为终端有限公司 | 图像拼接方法及装置 |
US20110169824A1 (en) * | 2008-09-29 | 2011-07-14 | Nobutoshi Fujinami | 3d image processing device and method for reducing noise in 3d image processing device |
US8233664B2 (en) * | 2008-11-12 | 2012-07-31 | Eastman Kodak Company | Determining relative depth of points in multiple videos |
CN103607585A (zh) * | 2009-01-21 | 2014-02-26 | 株式会社尼康 | 图像处理装置、图像处理方法 |
WO2010085549A1 (en) * | 2009-01-21 | 2010-07-29 | Vision Iii Imaging, Inc. | System and method for three-dimensional visualization of geographical data |
US9172940B2 (en) * | 2009-02-05 | 2015-10-27 | Bitanimate, Inc. | Two-dimensional video to three-dimensional video conversion based on movement between video frames |
CN101631256B (zh) * | 2009-08-13 | 2011-02-09 | 浙江大学 | 用于三维电视系统中2d视频到3d视频的转换方法 |
JP2011101230A (ja) * | 2009-11-06 | 2011-05-19 | Sony Corp | 表示制御装置、表示制御方法、プログラム、出力装置、および送信装置 |
WO2011063347A2 (en) | 2009-11-20 | 2011-05-26 | Pelican Imaging Corporation | Capturing and processing of images using monolithic camera array with heterogeneous imagers |
US9398289B2 (en) * | 2010-02-09 | 2016-07-19 | Samsung Electronics Co., Ltd. | Method and apparatus for converting an overlay area into a 3D image |
WO2011115142A1 (ja) * | 2010-03-19 | 2011-09-22 | Okiセミコンダクタ株式会社 | 画像処理装置、方法、プログラム及び記録媒体 |
US8928793B2 (en) | 2010-05-12 | 2015-01-06 | Pelican Imaging Corporation | Imager array interfaces |
US8421847B2 (en) * | 2010-05-21 | 2013-04-16 | Mediatek Inc. | Apparatus and method for converting two-dimensional video frames to stereoscopic video frames |
JP5530322B2 (ja) * | 2010-09-22 | 2014-06-25 | オリンパスイメージング株式会社 | 表示装置および表示方法 |
US8878950B2 (en) | 2010-12-14 | 2014-11-04 | Pelican Imaging Corporation | Systems and methods for synthesizing high resolution images using super-resolution processes |
TWI454130B (zh) * | 2010-12-28 | 2014-09-21 | Himax Media Solutions Inc | 二維轉三維的顏色補償系統及其方法 |
WO2012104759A1 (en) | 2011-02-04 | 2012-08-09 | Koninklijke Philips Electronics N.V. | Method of recording an image and obtaining 3d information from the image, camera system |
US9300947B2 (en) * | 2011-03-24 | 2016-03-29 | Kodak Alaris Inc. | Producing 3D images from captured 2D video |
WO2012155119A1 (en) | 2011-05-11 | 2012-11-15 | Pelican Imaging Corporation | Systems and methods for transmitting and receiving array camera image data |
EP2525581A3 (en) * | 2011-05-17 | 2013-10-23 | Samsung Electronics Co., Ltd. | Apparatus and Method for Converting 2D Content into 3D Content, and Computer-Readable Storage Medium Thereof |
TR201104918A2 (tr) | 2011-05-20 | 2012-12-21 | Vestel Elektroni̇k Sanayi̇ Ve Ti̇caret A.Ş. | Derinlik haritası ve 3d video oluşturmak için yöntem ve aygıt. |
US20130265459A1 (en) | 2011-06-28 | 2013-10-10 | Pelican Imaging Corporation | Optical arrangements for use with an array camera |
KR101888672B1 (ko) * | 2011-07-27 | 2018-08-16 | 엘지디스플레이 주식회사 | 입체영상 표시장치와 그 구동방법 |
JP5095850B1 (ja) * | 2011-08-31 | 2012-12-12 | 株式会社東芝 | オブジェクト探索装置、映像表示装置およびオブジェクト探索方法 |
WO2013043761A1 (en) | 2011-09-19 | 2013-03-28 | Pelican Imaging Corporation | Determining depth from multiple views of a scene that include aliasing using hypothesized fusion |
KR102002165B1 (ko) | 2011-09-28 | 2019-07-25 | 포토내이션 리미티드 | 라이트 필드 이미지 파일의 인코딩 및 디코딩을 위한 시스템 및 방법 |
IN2014CN03412A (ja) | 2011-11-01 | 2015-10-09 | Koninkl Philips Nv | |
JP5303692B1 (ja) * | 2011-11-28 | 2013-10-02 | パナソニック株式会社 | 立体画像処理装置及び立体画像処理方法 |
TWI483612B (zh) * | 2011-12-22 | 2015-05-01 | Nat Univ Chung Cheng | Converting the video plane is a perspective view of the video system |
DE102012100848B4 (de) | 2012-02-01 | 2019-05-29 | Hans Pfriem | System und Verfahren zur stereoskopischen Darstellung von Aufnahmen eines Endoskops |
US9412206B2 (en) | 2012-02-21 | 2016-08-09 | Pelican Imaging Corporation | Systems and methods for the manipulation of captured light field image data |
US9210392B2 (en) | 2012-05-01 | 2015-12-08 | Pelican Imaging Coporation | Camera modules patterned with pi filter groups |
CN104508681B (zh) | 2012-06-28 | 2018-10-30 | Fotonation开曼有限公司 | 用于检测有缺陷的相机阵列、光学器件阵列和传感器的系统及方法 |
US20140002674A1 (en) | 2012-06-30 | 2014-01-02 | Pelican Imaging Corporation | Systems and Methods for Manufacturing Camera Modules Using Active Alignment of Lens Stack Arrays and Sensors |
CN104662589B (zh) | 2012-08-21 | 2017-08-04 | 派力肯影像公司 | 用于使用阵列照相机捕捉的图像中的视差检测和校正的系统和方法 |
EP2888698A4 (en) | 2012-08-23 | 2016-06-29 | Pelican Imaging Corp | PROPERTY-BASED HIGH-RESOLUTION MOTION ESTIMATION FROM LOW-RESOLUTION IMAGES RECORDED WITH AN ARRAY SOURCE |
WO2014043641A1 (en) | 2012-09-14 | 2014-03-20 | Pelican Imaging Corporation | Systems and methods for correcting user identified artifacts in light field images |
US9250653B2 (en) | 2012-09-28 | 2016-02-02 | City University Of Hong Kong | Capturing, processing, and reconstructing audio and video content of mobile devices |
WO2014052974A2 (en) | 2012-09-28 | 2014-04-03 | Pelican Imaging Corporation | Generating images from light fields utilizing virtual viewpoints |
WO2014078443A1 (en) | 2012-11-13 | 2014-05-22 | Pelican Imaging Corporation | Systems and methods for array camera focal plane control |
WO2014130849A1 (en) | 2013-02-21 | 2014-08-28 | Pelican Imaging Corporation | Generating compressed light field representation data |
US9374512B2 (en) | 2013-02-24 | 2016-06-21 | Pelican Imaging Corporation | Thin form factor computational array cameras and modular array cameras |
WO2014138695A1 (en) | 2013-03-08 | 2014-09-12 | Pelican Imaging Corporation | Systems and methods for measuring scene information while capturing images using array cameras |
US8866912B2 (en) | 2013-03-10 | 2014-10-21 | Pelican Imaging Corporation | System and methods for calibration of an array camera using a single captured image |
US9521416B1 (en) | 2013-03-11 | 2016-12-13 | Kip Peli P1 Lp | Systems and methods for image data compression |
US9106784B2 (en) | 2013-03-13 | 2015-08-11 | Pelican Imaging Corporation | Systems and methods for controlling aliasing in images captured by an array camera for use in super-resolution processing |
WO2014164550A2 (en) | 2013-03-13 | 2014-10-09 | Pelican Imaging Corporation | System and methods for calibration of an array camera |
US9888194B2 (en) | 2013-03-13 | 2018-02-06 | Fotonation Cayman Limited | Array camera architecture implementing quantum film image sensors |
US9519972B2 (en) | 2013-03-13 | 2016-12-13 | Kip Peli P1 Lp | Systems and methods for synthesizing images from image data captured by an array camera using restricted depth of field depth maps in which depth estimation precision varies |
US9578259B2 (en) | 2013-03-14 | 2017-02-21 | Fotonation Cayman Limited | Systems and methods for reducing motion blur in images or video in ultra low light with array cameras |
US9100586B2 (en) | 2013-03-14 | 2015-08-04 | Pelican Imaging Corporation | Systems and methods for photometric normalization in array cameras |
WO2014150856A1 (en) | 2013-03-15 | 2014-09-25 | Pelican Imaging Corporation | Array camera implementing quantum dot color filters |
EP2973476A4 (en) | 2013-03-15 | 2017-01-18 | Pelican Imaging Corporation | Systems and methods for stereo imaging with camera arrays |
US10122993B2 (en) | 2013-03-15 | 2018-11-06 | Fotonation Limited | Autofocus system for a conventional camera that uses depth information from an array camera |
US9445003B1 (en) | 2013-03-15 | 2016-09-13 | Pelican Imaging Corporation | Systems and methods for synthesizing high resolution images using image deconvolution based on motion and depth information |
US9633442B2 (en) | 2013-03-15 | 2017-04-25 | Fotonation Cayman Limited | Array cameras including an array camera module augmented with a separate camera |
US9497429B2 (en) | 2013-03-15 | 2016-11-15 | Pelican Imaging Corporation | Extended color processing on pelican array cameras |
CA2909245C (en) * | 2013-04-09 | 2018-02-27 | Bitanimate, Inc. | Two-dimensional video to three-dimensional video conversion method and system |
WO2015048694A2 (en) | 2013-09-27 | 2015-04-02 | Pelican Imaging Corporation | Systems and methods for depth-assisted perspective distortion correction |
US9967546B2 (en) | 2013-10-29 | 2018-05-08 | Vefxi Corporation | Method and apparatus for converting 2D-images and videos to 3D for consumer, commercial and professional applications |
US20150116458A1 (en) | 2013-10-30 | 2015-04-30 | Barkatech Consulting, LLC | Method and apparatus for generating enhanced 3d-effects for real-time and offline appplications |
EP3066690A4 (en) | 2013-11-07 | 2017-04-05 | Pelican Imaging Corporation | Methods of manufacturing array camera modules incorporating independently aligned lens stacks |
US10119808B2 (en) | 2013-11-18 | 2018-11-06 | Fotonation Limited | Systems and methods for estimating depth from projected texture using camera arrays |
WO2015081279A1 (en) | 2013-11-26 | 2015-06-04 | Pelican Imaging Corporation | Array camera configurations incorporating multiple constituent array cameras |
WO2015134996A1 (en) | 2014-03-07 | 2015-09-11 | Pelican Imaging Corporation | System and methods for depth regularization and semiautomatic interactive matting using rgb-d images |
CN103824303A (zh) * | 2014-03-14 | 2014-05-28 | 格科微电子(上海)有限公司 | 基于被摄物的位置、方向调整图像透视畸变的方法和装置 |
US9247117B2 (en) | 2014-04-07 | 2016-01-26 | Pelican Imaging Corporation | Systems and methods for correcting for warpage of a sensor array in an array camera module by introducing warpage into a focal plane of a lens stack array |
WO2015191767A1 (en) * | 2014-06-10 | 2015-12-17 | Bitanimate, Inc. | Stereoscopic depth adjustment and focus point adjustment |
US9591290B2 (en) * | 2014-06-10 | 2017-03-07 | Bitanimate, Inc. | Stereoscopic video generation |
US9693042B2 (en) | 2014-06-10 | 2017-06-27 | Bitanimate, Inc. | Foreground and background detection in a video |
US9661309B2 (en) * | 2014-06-10 | 2017-05-23 | Bitanimate, Inc. | Stereoscopic video zooming |
WO2015191768A1 (en) * | 2014-06-10 | 2015-12-17 | Bitanimate, Inc | Stereoscopic video zooming and foreground and background detection in a video |
AU2014397095B2 (en) * | 2014-06-12 | 2018-07-05 | Eizo Corporation | Haze removal device and image generation method |
US9521319B2 (en) | 2014-06-18 | 2016-12-13 | Pelican Imaging Corporation | Array cameras and array camera modules including spectral filters disposed outside of a constituent image sensor |
US10158847B2 (en) | 2014-06-19 | 2018-12-18 | Vefxi Corporation | Real—time stereo 3D and autostereoscopic 3D video and image editing |
EP3201877B1 (en) | 2014-09-29 | 2018-12-19 | Fotonation Cayman Limited | Systems and methods for dynamic calibration of array cameras |
US9942474B2 (en) | 2015-04-17 | 2018-04-10 | Fotonation Cayman Limited | Systems and methods for performing high speed video capture and depth estimation using array cameras |
DE102015112411A1 (de) | 2015-07-29 | 2017-02-02 | Universal Consulting GmbH | System zur stereoskopischen Darstellung von Aufnahmen eines Endoskops |
US10277879B2 (en) | 2015-11-13 | 2019-04-30 | Vefxi Corporation | 3D system including rendering with eye displacement |
US10242448B2 (en) | 2015-11-13 | 2019-03-26 | Vefxi Corporation | 3D system including queue management |
US10277880B2 (en) | 2015-11-13 | 2019-04-30 | Vefxi Corporation | 3D system including rendering with variable displacement |
US10148933B2 (en) | 2015-11-13 | 2018-12-04 | Vefxi Corporation | 3D system including rendering with shifted compensation |
US10284837B2 (en) | 2015-11-13 | 2019-05-07 | Vefxi Corporation | 3D system including lens modeling |
US10121280B2 (en) | 2015-11-13 | 2018-11-06 | Vefxi Corporation | 3D system including rendering with three dimensional transformation |
WO2017083509A1 (en) * | 2015-11-13 | 2017-05-18 | Craig Peterson | Three dimensional system |
US10277877B2 (en) | 2015-11-13 | 2019-04-30 | Vefxi Corporation | 3D system including a neural network |
US10148932B2 (en) | 2015-11-13 | 2018-12-04 | Vefxi Corporation | 3D system including object separation |
US20170140571A1 (en) * | 2015-11-13 | 2017-05-18 | Craig Peterson | 3d system including rendering with curved display |
US10122987B2 (en) | 2015-11-13 | 2018-11-06 | Vefxi Corporation | 3D system including additional 2D to 3D conversion |
US10225542B2 (en) | 2015-11-13 | 2019-03-05 | Vefxi Corporation | 3D system including rendering with angular compensation |
US10154244B2 (en) | 2016-02-18 | 2018-12-11 | Vefxi Corporation | 3D system including a marker mode |
US10482618B2 (en) | 2017-08-21 | 2019-11-19 | Fotonation Limited | Systems and methods for hybrid depth regularization |
CN110673114B (zh) * | 2019-08-27 | 2023-04-18 | 三赢科技(深圳)有限公司 | 校准三维相机深度的方法、装置、计算机装置及存储介质 |
WO2021055585A1 (en) | 2019-09-17 | 2021-03-25 | Boston Polarimetrics, Inc. | Systems and methods for surface modeling using polarization cues |
MX2022004163A (es) | 2019-10-07 | 2022-07-19 | Boston Polarimetrics Inc | Sistemas y metodos para la deteccion de estandares de superficie con polarizacion. |
MX2022005289A (es) | 2019-11-30 | 2022-08-08 | Boston Polarimetrics Inc | Sistemas y metodos para segmentacion de objetos transparentes usando se?ales de polarizacion. |
WO2021154386A1 (en) | 2020-01-29 | 2021-08-05 | Boston Polarimetrics, Inc. | Systems and methods for characterizing object pose detection and measurement systems |
US11797863B2 (en) | 2020-01-30 | 2023-10-24 | Intrinsic Innovation Llc | Systems and methods for synthesizing data for training statistical models on different imaging modalities including polarized images |
WO2021243088A1 (en) | 2020-05-27 | 2021-12-02 | Boston Polarimetrics, Inc. | Multi-aperture polarization optical systems using beam splitters |
US11290658B1 (en) | 2021-04-15 | 2022-03-29 | Boston Polarimetrics, Inc. | Systems and methods for camera exposure control |
US11954886B2 (en) | 2021-04-15 | 2024-04-09 | Intrinsic Innovation Llc | Systems and methods for six-degree of freedom pose estimation of deformable objects |
US11689813B2 (en) | 2021-07-01 | 2023-06-27 | Intrinsic Innovation Llc | Systems and methods for high dynamic range imaging using crossed polarizers |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1198527A (ja) * | 1997-09-18 | 1999-04-09 | Sanyo Electric Co Ltd | 2次元映像を3次元映像に変換する装置および方法 |
JPH1198530A (ja) * | 1997-09-22 | 1999-04-09 | Sanyo Electric Co Ltd | 2次元映像を3次元映像に変換する装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100358021B1 (ko) | 1994-02-01 | 2003-01-24 | 산요 덴키 가부시키가이샤 | 2차원영상을3차원영상으로변환시키는방법및입체영상표시시스템 |
JP3214688B2 (ja) * | 1994-02-01 | 2001-10-02 | 三洋電機株式会社 | 2次元映像を3次元映像に変換する方法及び3次元映像信号生成装置 |
JP2846830B2 (ja) * | 1994-11-22 | 1999-01-13 | 三洋電機株式会社 | 2次元映像を3次元映像に変換する方法 |
AUPO894497A0 (en) * | 1997-09-02 | 1997-09-25 | Xenotech Research Pty Ltd | Image processing method and apparatus |
-
2000
- 2000-11-24 AT AT00977933T patent/ATE278298T1/de not_active IP Right Cessation
- 2000-11-24 DE DE60014420T patent/DE60014420T2/de not_active Expired - Lifetime
- 2000-11-24 TW TW089125162A patent/TW521519B/zh not_active IP Right Cessation
- 2000-11-24 US US10/130,427 patent/US7161614B1/en not_active Expired - Fee Related
- 2000-11-24 EP EP00977933A patent/EP1235438B1/en not_active Expired - Lifetime
- 2000-11-24 WO PCT/JP2000/008315 patent/WO2001039512A1/ja active IP Right Grant
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1198527A (ja) * | 1997-09-18 | 1999-04-09 | Sanyo Electric Co Ltd | 2次元映像を3次元映像に変換する装置および方法 |
JPH1198530A (ja) * | 1997-09-22 | 1999-04-09 | Sanyo Electric Co Ltd | 2次元映像を3次元映像に変換する装置 |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8369607B2 (en) | 2002-03-27 | 2013-02-05 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8472702B2 (en) | 2002-03-27 | 2013-06-25 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8879824B2 (en) | 2002-03-27 | 2014-11-04 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8131064B2 (en) | 2002-03-27 | 2012-03-06 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8724886B2 (en) | 2002-03-27 | 2014-05-13 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8254668B2 (en) | 2002-03-27 | 2012-08-28 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
EP1489857A4 (en) * | 2002-03-27 | 2007-07-25 | Sanyo Electric Co | THREE-DIMENSIONAL IMAGE PROCESSING METHOD AND DEVICE |
EP1489857A1 (en) * | 2002-03-27 | 2004-12-22 | Sanyo Electric Co., Ltd. | 3-dimensional image processing method and device |
US8577127B2 (en) | 2002-03-27 | 2013-11-05 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8559703B2 (en) | 2002-03-27 | 2013-10-15 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8417024B2 (en) | 2002-03-27 | 2013-04-09 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
US8577128B2 (en) | 2002-03-27 | 2013-11-05 | Sanyo Electric Co., Ltd. | Method and apparatus for processing three-dimensional images |
CN101815225A (zh) * | 2009-02-25 | 2010-08-25 | 三星电子株式会社 | 生成深度图的方法及其设备 |
CN102427550A (zh) * | 2011-12-09 | 2012-04-25 | 彩虹集团公司 | 一种实现2d转3d的方法 |
Also Published As
Publication number | Publication date |
---|---|
EP1235438A4 (en) | 2003-03-05 |
EP1235438B1 (en) | 2004-09-29 |
ATE278298T1 (de) | 2004-10-15 |
TW521519B (en) | 2003-02-21 |
DE60014420T2 (de) | 2005-10-13 |
DE60014420D1 (de) | 2004-11-04 |
US7161614B1 (en) | 2007-01-09 |
EP1235438A1 (en) | 2002-08-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2001039512A1 (en) | Device and method for converting two-dimensional video to three-dimensional video | |
JP2001320731A (ja) | 2次元映像を3次元映像に変換する装置及びその方法 | |
KR100445619B1 (ko) | 2차원 영상을 3차원 영상으로 변환하는 장치 및 방법 | |
US6466255B1 (en) | Stereoscopic video display method and apparatus, stereoscopic video system, and stereoscopic video forming method | |
US8472704B2 (en) | Image processing apparatus and image processing method | |
KR100414629B1 (ko) | 3차원표시화상생성방법,깊이정보를이용한화상처리방법,깊이정보생성방법 | |
US8687041B2 (en) | Stereoscopic panorama image creating apparatus, stereoscopic panorama image creating method, stereoscopic panorama image reproducing apparatus, stereoscopic panorama image reproducing method, and recording medium | |
KR101856805B1 (ko) | 화상 처리 장치 및 화상 처리 방법, 및 프로그램 | |
US20130215107A1 (en) | Image processing apparatus, image processing method, and program | |
JP2012257022A (ja) | 画像処理装置および方法、並びにプログラム | |
JP2012138787A (ja) | 画像処理装置、および画像処理方法、並びにプログラム | |
US20120026165A1 (en) | Image Processing Apparatus and Method, and Program | |
KR20110071528A (ko) | 스테레오 영상, 다시점 영상 및 깊이 영상 획득 카메라 장치 및 그 제어 방법 | |
US20130293533A1 (en) | Image processing apparatus and image processing method | |
JP5669599B2 (ja) | 画像処理装置及びその制御方法 | |
KR100924432B1 (ko) | 다시점 영상의 인식 깊이감 조절 장치 및 방법 | |
US9386291B2 (en) | Video signal processing device | |
US9210396B2 (en) | Stereoscopic image generation apparatus and stereoscopic image generation method | |
Kim et al. | Stereoscopic conversion of monoscopic video by the transformation of vertical-to-horizontal disparity | |
JP3197801B2 (ja) | 二次元表示画像生成方法 | |
JP5127973B1 (ja) | 映像処理装置、映像処理方法および映像表示装置 | |
JP4036599B2 (ja) | 2次元映像を3次元映像に変換する方法 | |
JP3540626B2 (ja) | 2次元映像を3次元映像に変換する装置および方法 | |
KR101329069B1 (ko) | 깊이 추정 데이터 생성 장치, 깊이 추정 데이터 생성 프로그램이 기록된 컴퓨터 판독 가능한 기록 매체 및, 의사 입체 화상 표시 장치 | |
KR20140119999A (ko) | 깊이값 조정을 통한 입체영상 제공 시스템 및 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 10130427 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2000977933 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2000977933 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 2000977933 Country of ref document: EP |