CN103164690A - Method and device for utilizing motion tendency to track augmented reality three-dimensional multi-mark - Google Patents
Method and device for utilizing motion tendency to track augmented reality three-dimensional multi-mark Download PDFInfo
- Publication number
- CN103164690A CN103164690A CN2011104261308A CN201110426130A CN103164690A CN 103164690 A CN103164690 A CN 103164690A CN 2011104261308 A CN2011104261308 A CN 2011104261308A CN 201110426130 A CN201110426130 A CN 201110426130A CN 103164690 A CN103164690 A CN 103164690A
- Authority
- CN
- China
- Prior art keywords
- mark
- image
- camera image
- identified
- frame
- 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
Links
Images
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention provides a method and a device for utilizing a motion tendency to track augmented reality three-dimensional multi marks. The method for utilizing the motion tendency to track the augmented reality three-dimensional multi marks comprises the following steps: receiving a frame camera image; identifying a result according to a position of a mark existing in a front frame in the frame camera image, and generating a motion tendency of the mark; dynamically ensuring a sequence of identification of regions in the frame camera image according to the motion tendency; and identifying the regions in the frame camera image according to the sequence, and inserting a three-dimensional virtual article on an identified position when the mark is identified in one region of the frame camera image.
Description
Technical field
Relate generally to field of image recognition of the present invention more specifically, the present invention relates to utilize movement tendency to follow the tracks of the method and apparatus of the three-dimensional multiple labeling of augmented reality (Augmented Reality is called for short AR).
Background technology
Augmented reality is a kind of technology of calculating in real time position and the angle of camera image and adding respective image, and the target of this technology is on screen, virtual world to be enclosed within real world and to carry out interaction.Along with the lifting of accompanied electronic product arithmetic capability, the purposes of expection augmented reality will be more and more wider.The Ronald Azuma of North Carolina University proposed the content that augmented reality comprises three aspects in 1997: with virtual object and reality combination, immediate interactive and three-dimensional (3D).
At present augmented reality mainly contains following several implementation: (1) by GPS (GPS), geomagnetic sensor and acceleration transducer determine user terminal the geographic position, towards and angle of inclination etc., then obtain Overlapping display after relevant information according to determined positional information, adopt this implementation such as PlaceEngine technology, SekaiCamera technology etc. are arranged; (2) information of pre-save mark (marker) image, then pass through image recognition technology, search in present image and the identification marking image, the relevant information that then superposes on marking image, adopt this implementation such as the ARToolKit routine library is arranged; And resolve photographs (3), identify landscape, object and space, then relevant information superposes, adopt this implementation such as have proposed by the Georg Klein of Regius professor and David Murray and line trace and mapping (Parallel Tracking and Mapping is called for short PTAM) development library.
Among mentioned implementation, by mark being identified realize that the technology of information stack is used more and more widely.This implementation can be by identifying to add virtual 3D object to the mark in true picture with video camera, thereby cause the visual effect of actual situation combination.Yet when needs were identified a plurality of mark simultaneously, recognizer need to judge for the unique point (feature point) of each mark.Except AR Toolkit, recognition methods commonly used at present also has Institute Graphische Datenverarbeitung, SCR (Semans AG), Hoffman Marker System (HMS) etc.
For the image that is photographed by camera lens, we must manage to identify special marking wherein, and with the direction of this mark, the 3D virtual article of coming the insertion program to produce.
In the application of a plurality of movement marks of needs identification, the efficient of identification becomes important consideration.If the efficient of identification is too low, can cause 3D virtual article having some setbacks in motion process.
Mark in the identification camera image, mainly use the judging characteristic value to calculate the method for so-called degree of confidence (confidence level) at present, when degree of confidence surpasses certain numerical value, just is considered as detecting mark.Yet the judgement of eigenwert must judge for individual other mark.When the reference numerals quantitative change was many, therefore the time of judgement also increased.
Tradition AR 3D carries out the identification of a plurality of 3D marks for the current camera image that captures.Present recognition methods is identified for single mark mostly, or identifies in limited quantity (such as 2-3 's) mark.When this identifies a plurality of mark simultaneously at needs, can cause a large amount of image to process and calculate, thereby can't reach the requirement of real-time calculating and identification on the limited mobile platform of the counting yielies such as handheld device.
Summary of the invention
Consider the problems referred to above, the present invention proposes the efficient that a kind of method and apparatus that utilizes movement tendency to follow the tracks of the three-dimensional multiple labeling of augmented reality improves a plurality of marks of AR 3D identification.The present invention uses the continuity of 3D object motion between consecutive image, accelerates identification and tracking to a plurality of 3D marks in image.
According to an aspect of the present invention, a kind of mark recognition method comprises the following steps: receive a frame camera image; According to a position recognition result that is marked in front frame in described image, produce the movement tendency of a described mark; According to described movement tendency, dynamically determine the order that the regional in this frame camera image is identified; And by described order, the regional in this frame camera image is identified, and when identifying a described mark in a zone in this frame camera image, insert the three-dimensional object in the position of identifying.
The method is considered the continuity of object motion, has dwindled for the identification scanning area that is marked at individually on the consecutive image that video camera imports into, therefore can when the 3D number of labels increases, reduce the computation complexity of identifying each mark.When the method increases at the marker number that will identify, can less than to time that marker number is directly proportional in all marks in identifying image.
In addition, therefore the method can find out the reposition that is marked in image than existing recognition methods more quickly because the identification history of usage flag reduces the identified region that needs comparison.
In the method according to the invention, the step of generation movement tendency can comprise: according to a position recognition result that is marked in front frame, determine that this is marked at the possible position that occurs in current camera image frame.Current camera image frame can be divided into m * n image block areas, and the determined order of identifying can be: the first image block areas to the possible position place is identified; Eight second image block areas adjacent with the first image block areas are identified; And other image block areas in this frame camera image are identified, and m and n can be the integers greater than 1.Can dynamically determine according to movement tendency the order that eight the second image block areas are identified.Determining and to complete by the linear extrapolation algorithm possible position.
In the method according to the invention, described 2 frames before front frame can be the frame camera image that receives of next-door neighbour.
In the method according to the invention, a frame camera image can comprise 3 or more mark.
According to a further aspect in the invention, a kind of mark recognition device comprises: receiving trap is used for receiving a frame camera image; The movement tendency generation device is used for a position recognition result that is marked in front frame according to described image, produces the movement tendency of a described mark; Recognition sequence is determined device, is used for according to described movement tendency, dynamically determines the order that the regional in this frame camera image is identified; And three-dimensional object insertion apparatus, be used for by described order, the regional of this frame camera image being identified, and when identifying a described mark in a zone in this frame camera image, insert the three-dimensional object in the position of identifying.
According to another aspect of the invention, a kind of mark identification terminal equipment comprises above-mentioned mark recognition device.For example, this mark terminal device can be the equipment such as mobile phone, PDA, panel computer, laptop computer, desktop PC.
The method according to this invention and device are by preferentially identifying the zone that more may occur mark in the reception image, can reduce the calculated amount of mark identification, thus identify rapidly mark with guarantee in real time identification marking and with the 3D virtual article be inserted into received in image.By improving the efficient that mark is identified, even in the situation that needs are identified a plurality of marks (such as 3 or more mark) simultaneously, method and apparatus of the present invention also can be guaranteed in real time identification marking and insert the 3D virtual article.
From the detailed description below in conjunction with accompanying drawing, can find out other features and advantages of the present invention.Note, the present invention is not limited to the example shown in figure or any specific embodiment.
Description of drawings
By reference to the accompanying drawings, from following detailed description to the embodiment of the present invention, will understand better the present invention, similarly indicate similar part with reference to mark in accompanying drawing, wherein:
Fig. 1 is that the zone to a two field picture that according to a present invention concrete example is shown is divided and for the diagram of a mark to the identification scanning sequency in each zone in this two field picture;
Fig. 2 is that the zone to a two field picture that according to the present invention another concrete example is shown is divided and for the diagram of a mark to the identification scanning sequency in each zone in this two field picture;
Fig. 3 is the process flow diagram that the flow process of the method for utilizing the three-dimensional multiple labeling of movement tendency tracking augmented reality of a concrete example according to the present invention is shown; With
Fig. 4 is the block diagram that the inner structure of the device that utilizes the three-dimensional multiple labeling of movement tendency tracking augmented reality of a concrete example according to the present invention is shown.
Embodiment
A lot of details have been set forth below in detailed description of the present invention, so that fully understand the present invention.But, do not have these details can implement the present invention yet, be clearly for a person skilled in the art.In the other example, known method, process, parts and circuit are not described in detail, a presumptuous guest usurps the role of the host to avoid, desalinated main contents of the present invention.
Below, provide a concrete example of the present invention.
Fig. 1 shows to the zone division of a frame camera image and for the order of identifying scanning of a mark to the regional in this two field picture.
As shown in Figure 1, a frame camera image that receives is divided into 4 * 3 image block areas,, represents one of them image block with (x, y) here, and 0＜=x＜4 and 0＜=y＜3.
In the time will beginning one of a plurality of marks in camera image are as shown in Figure 1 identified, from routine to its carry out from left to right, from top to bottom scanning recognition is different, at first method of the present invention is marked at position recognition result in front frame based on that will identify, produces the movement tendency of this mark.For example can with the next-door neighbour of this mark of preserving the preceding the position recognition result in 2 two field pictures calculate its movement tendency.Particularly, can predict that this is marked at the possible position that occurs in this frame camera image with the linear extrapolation algorithm:
(x wherein
k, y
k) be the last inferior position that is marked on two-dimensional image frame, and (x
k-1, y
k-1) be that more the front once is marked at position on two-dimensional image frame.Utilize these two near possible position (x
*, y
*) the first two recognizing site, can dope the possible position (x that the mark most probable occurs
*, y
*).As shown in Figure 1, the movement tendency that has been shown in dotted line this mark of the position in former frame of linkage flag (last position) and possible position.Perhaps in other words, represented the movement tendency of this mark take the last position of mark as the vector (vector) of terminal point as starting point and take the possible position of mark.Below this vector is called the movement tendency vector.
Then, according to the movement tendency that produces, dynamically determine the order that each image block areas in this frame camera image is identified.In the example depicted in fig. 1, image block (the x+1 at the possible position place that at first occurs at the mark most probable, y+1) identify in, then with the more closely-related image block (x of movement tendency, y+1) and (x+1, y) identify in, identify in image block (x, y) at last.As shown in Figure 1, the selecting sequence of identified region is the order of numbering 1,2,3 and 4.If all can not recognize this mark in being numbered 1,2,3 and 4 image block, then in eight image blocks around this possible position, identify in the image block not yet searched.If still do not recognize this mark, again the residual image piece in this two field picture is identified.Alternately, when determining the image block at possible position place, the movement tendency vector can be converted to the side-play amount on horizontal direction and vertical direction.In the example depicted in fig. 1, the off-set value after movement tendency vector conversion is (1,1), so the image block at possible position place is (x+1, y+1).
That is to say, recognition sequence determined according to movement tendency as shown in Figure 1 is such: at first 1 the image block of being numbered at possible position place is identified, then to be numbered among 8 image blocks around 1 image block, be numbered 2,3,4 image block and identify successively, then identify being numbered image block among 8 image blocks around 1 image block, except being numbered 2,3,4 image block again, at last the residual image piece in this two field picture is identified.
Next, according to the order of determining as mentioned above, the regional in camera image is identified to identify mark.When identifying the position of mark in a zone in camera image, insert corresponding three-dimensional object according to the direction of the position of identifying and this position.Particularly, if identified mark in being numbered 1 image block, no longer continue identifying, but insert as required suitable 3D virtual article in the marked locations that identifies.If unidentifiedly in being numbered 1 image block go out mark, continue to carry out identification scanning according to the order of determining as mentioned above, in case identify mark insert the 3D virtual article, otherwise just continue in order scanning recognition.To be numbered order that image block among 8 image blocks around 1 image block, except being numbered 2,3,4 image block identifies and to the order that the residual image piece in image is identified can be from left to right, from top to bottom order.
In example illustrated in fig. 1, historical by the identification of usage flag, the zone that can preferentially more may occur mark is identified, thereby has reduced the computation complexity that mark is identified.Therefore, compare with existing mark recognition method, method of the present invention can find the reposition that is marked in a frame camera image more quickly.Especially in the time will identifying simultaneously 3 or more mark, the present invention can less than to time that marker number is directly proportional in all marks in identifying image.
Fig. 2 show according to another example to the zone of a frame camera image dividing and for the order of identifying scanning of a mark to the regional in this two field picture.Except the direction difference of movement tendency, situation illustrated in fig. 2 is identical with situation illustrated in fig. 1.
when the direction of movement tendency changes, as shown in Figure 2, when movement tendency is from left to right the time, the order that each image block in camera image is identified of determining according to this movement tendency can be: at first 1 the image block of being numbered at possible position place is searched for, then to be numbered 1 image block below be numbered 2 image blocks with and 3 image blocks that are numbered in left side search for successively, then to being numbered among 8 image blocks around 1 image block, except being numbered 2, image block outside 3 image block is identified, at last the residual image piece in camera image is identified.
Can find out from example shown in Figure 2, according to the difference of the direction of movement tendency, the order that eight image blocks around the image block at possible position place are identified also is not quite similar.Preferably, the vector of the close movement tendency among these eight image blocks of first search or more relevant those image blocks with it.
Below, be described with reference to Figure 3 and utilize movement tendency to follow the tracks of the idiographic flow of the method for the three-dimensional multiple labeling of augmented reality according to the present invention.The process flow diagram of Fig. 3 shows the method step under situation shown in Fig. 1.At step S201, receive a frame camera image.Due to as mentioned above, determine movement tendency by the mark position recognition result in front image based at least 2 frames, so the camera image that receives here should be the 3rd frame camera image or camera image afterwards.At step S202, this frame camera image that receives is divided into 4 * 3 image block areas.At step S203, be marked at the position recognition result that is right after in 2 frames the preceding based on that will identify, produce the movement tendency of this mark.This generation step comprises based on the position recognition result that is right after in 2 frames the preceding, predicts that with the linear extrapolation algorithm this is marked at the possible position that occurs in this frame camera image.Take the last position of mark as starting point and take mark the possible position of being predicted represented the movement tendency of this mark as the vector of terminal point.At step S204, judge whether to have identified mark in being numbered 1 image block.If so, method flow advances to step S207, at this step place, inserts the three-dimensional object corresponding with mark according to the direction of the position of identifying and this position.If unidentifiedly in being numbered 1 image block go out mark, method flow advances to step S205, at this step place, judges whether to go out mark at 8 image recognitions that are numbered around 1 image block.Particularly, in the situation that shown in Figure 1, identify successively being numbered 2,3 and 4 image block, then other image blocks in these 8 image blocks are identified by from top to bottom order from left to right.If identified mark in an image block in these 8 image blocks, method flow advances to step S207.If all unidentifiedly in the arbitrary image block in these 8 image blocks go out mark, method flow advances to step S206, at step S206 place, judges whether to identify mark in other image blocks in this frame camera image.If search mark in one in these residual image pieces, method flow advances to step S207.If all do not search mark in any in these residual image pieces, illustrate that mark may shift out camera lens or mark is hidden or to the mark recognition failures, in the case, method flow advances to step S208, at this step place, judge whether to also have next mark to be identified.If also have mark to identify, method flow turns back to step S203, and then the processing from step S203 to step S208 is repeated.If handled all marks, method flow finishes.
Fig. 4 illustrates the block diagram of the inner structure of the device 400 that utilizes the three-dimensional multiple labeling of movement tendency tracking augmented reality of a concrete example according to the present invention.As shown in the figure, this device 400 comprises that receiving trap 401, movement tendency generation device 402, recognition sequence determine device 403 and three-dimensional object insertion apparatus 404.Receiving trap 401 receives a frame camera image, and this frame camera image is outputed to movement tendency generation device 402.Movement tendency generation device 402 produces the movement tendency of a described mark according to a position recognition result that is marked in front frame in this two field picture, and this movement tendency is outputed to recognition sequence determines device 403.Recognition sequence is determined device 403 according to the movement tendency that receives, and dynamically determines order that the regional in this frame camera image is identified, and with determined Sequential output to three-dimensional object insertion apparatus 404.Three-dimensional object insertion apparatus 404 is identified the regional in this frame camera image by determined order, and when identifying a described mark in a zone in this frame camera image, insert the three-dimensional object in the position of identifying.
The above has been described with reference to the drawings according to concrete example of the present invention.But the present invention is not limited to the particular condition shown in figure and processing.
For example, in the concrete example that provides, when producing the movement tendency of mark, used next-door neighbour's position recognition result in 2 two field pictures the preceding.Yet, obviously also can use any two position recognition result in prior image frame, perhaps use more position recognition result in prior image frame, realize same even better effect.
In addition, in the concrete example that provides, calculate the possible position of mark with the linear extrapolation algorithm.Yet this algorithm is only exemplary.Can also use hereby (Bezier) curve etc. possible position of coming predictive marker to occur of shellfish.
In addition, in the concrete example that provides, a two field picture is divided into 4 * 3 image block areas, but obviously also can according to the actual requirements, a two field picture be divided into more or less image block areas.For example, a two field picture can be divided into 3 * 3,3 * 4,8 * 6 even 8 * 8 image blocks, etc.This depends on the number of the pixel aspect ratio of two field picture, the mark that will identify, desired frame rate etc. usually.
In addition, in the concrete example that provides, a two field picture is divided into the individual image block areas of m * n (m=4, n=3), yet such division is not essential.For example, in a concrete example, in the situation that determined the possible position of a mark most probable in appearing at a two field picture, can be preferentially a certain presumptive area at this possible position place be searched for, and then the zone except this presumptive area in this two field picture is searched for.This presumptive area can be the square image-region centered by determined possible position, and the length of side of this square image-region is the length or wide 1/2nd, 1/3rd etc. of this two field picture.Perhaps, the length of side of this square image-region is a predetermined value.
In addition, in the concrete example that provides, according to the difference of movement tendency vector direction, dynamically determine the order that image each zone is identified.Yet, also can dynamically determine according to the difference of movement tendency vector magnitude the order that image each zone is identified.
In the above-described embodiments, describe and show some concrete steps as example.But procedure of the present invention is not limited to the concrete steps that institute describes and illustrates, and those skilled in the art can make various changes, modification and interpolation after understanding spirit of the present invention, perhaps change the order between step.
Element of the present invention can be implemented as hardware, software, firmware or their combination, and can be used in their system, subsystem, parts or subassembly.When realizing with software mode, element of the present invention is program or the code segment that is used to carry out required task.Program or code segment can be stored in machine readable media, perhaps send at transmission medium or communication links by the data-signal that carries in carrier wave." machine readable media " can comprise and can store or any medium of transmission information.The example of machine readable media comprises electronic circuit, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disk, CD-ROM, CD, hard disk, fiber medium, radio frequency (RF) link, etc.Code segment can be downloaded via the computer network such as the Internet, Intranet etc.
The present invention can realize with other concrete form, and do not break away from its spirit and essential characteristic.Therefore, current embodiment is counted as exemplary but not determinate in all respects, scope of the present invention is by claims but not foregoing description definition, and, thereby the whole changes that fall in the scope of the implication of claim and equivalent all are included among scope of the present invention.
Claims (15)
1. mark recognition method comprises:
Receive a frame camera image;
According to a position recognition result that is marked in front frame in described image, produce the movement tendency of a described mark;
According to described movement tendency, dynamically determine the order that the regional in this frame camera image is identified; And
By described order, the regional in this frame camera image is identified, and when identifying a described mark in a zone in this frame camera image, inserted the three-dimensional object in the position of identifying.
2. method according to claim 1, wherein, the step of described generation movement tendency comprises: according to a described position recognition result that is marked in front frame, determine the described possible position that occurs in this frame camera image that is marked at.
3. method according to claim 2, wherein, this frame camera image is divided into m * n image block areas, and the determined order of identifying is:
The first image block areas to described possible position place is identified;
Eight second image block areas adjacent with described the first image block are identified; And
Other image block areas in this frame camera image are identified,
And wherein, m and n are the integers greater than 1.
4. method according to claim 3, wherein, dynamically determine according to described movement tendency the order that described eight the second image block areas are identified.
5. method according to claim 2, wherein, to described possible position determine complete by the linear extrapolation algorithm.
6. method according to claim 1, wherein, described 2 frames before front frame is the frame camera image that receives of next-door neighbour.
7. method according to claim 1, wherein, comprise 3 or more mark in described camera image.
8. mark recognition device comprises:
Receiving trap is used for receiving a frame camera image;
The movement tendency generation device is used for a position recognition result that is marked in front frame according to described image, produces the movement tendency of a described mark;
Recognition sequence is determined device, is used for according to described movement tendency, dynamically determines the order that the regional in this frame camera image is identified; And
Three-dimensional object insertion apparatus, be used for by described order, the regional of this frame camera image being identified, and when identifying a described mark in a zone in this frame camera image, insert the three-dimensional object in the position of identifying.
9. device according to claim 8, wherein, described movement tendency generation device comprises: possible position is determined device, is used for according to a described position recognition result that is marked at front frame, determines the described device that is marked at the possible position that occurs in this frame camera image.
10. device according to claim 9, wherein, this frame camera image is divided into m * n image block areas, and the determined order of identifying is:
The first image block areas to described possible position place is identified;
Eight second image block areas adjacent with described the first image block are identified; And
Other image block areas in this frame camera image are identified,
And wherein, m and n are the integers greater than 1.
11. device according to claim 10 wherein, is dynamically determined according to described movement tendency the order that described eight the second image block areas are identified.
12. device according to claim 9, wherein, to described possible position determine complete by the linear extrapolation algorithm.
13. device according to claim 8, wherein, described 2 frames before front frame is the frame camera image that receives of next-door neighbour.
14. device according to claim 8 wherein, comprises 3 or more mark in described camera image.
15. a mark identification terminal equipment comprises: any one described device according to claim 8-14.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104261308A CN103164690A (en) | 2011-12-09 | 2011-12-09 | Method and device for utilizing motion tendency to track augmented reality three-dimensional multi-mark |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104261308A CN103164690A (en) | 2011-12-09 | 2011-12-09 | Method and device for utilizing motion tendency to track augmented reality three-dimensional multi-mark |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103164690A true CN103164690A (en) | 2013-06-19 |
Family
ID=48587762
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011104261308A Pending CN103164690A (en) | 2011-12-09 | 2011-12-09 | Method and device for utilizing motion tendency to track augmented reality three-dimensional multi-mark |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103164690A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016005948A3 (en) * | 2014-07-11 | 2016-05-26 | Idvision Limited | Augmented reality system |
| CN106096540A (en) * | 2016-06-08 | 2016-11-09 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN107403441A (en) * | 2016-05-19 | 2017-11-28 | 视辰信息科技(上海)有限公司 | The tracking and terminal device of augmented reality system |
| CN110248127A (en) * | 2019-06-20 | 2019-09-17 | 陕西科技大学 | A kind of broadcast transmitter radio frequency unit signal amplitude modulation circuit |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101197647A (en) * | 2006-12-13 | 2008-06-11 | 四川川大智胜软件股份有限公司 | Multi-channel real-time three-dimensional vision rendering indication method |
| CN101344920A (en) * | 2008-07-21 | 2009-01-14 | 北大方正集团有限公司 | Method and device for detecting specific area in video data frame |
| CN102110379A (en) * | 2011-02-22 | 2011-06-29 | 黄振强 | Multimedia reading matter giving readers enhanced feeling of reality |
| WO2011078430A1 (en) * | 2009-12-22 | 2011-06-30 | (주)포스트미디어 | Sequential searching method for recognizing plurality of feature point-based markers and augmented reality implementing method using same |
-
2011
- 2011-12-09 CN CN2011104261308A patent/CN103164690A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101197647A (en) * | 2006-12-13 | 2008-06-11 | 四川川大智胜软件股份有限公司 | Multi-channel real-time three-dimensional vision rendering indication method |
| CN101344920A (en) * | 2008-07-21 | 2009-01-14 | 北大方正集团有限公司 | Method and device for detecting specific area in video data frame |
| WO2011078430A1 (en) * | 2009-12-22 | 2011-06-30 | (주)포스트미디어 | Sequential searching method for recognizing plurality of feature point-based markers and augmented reality implementing method using same |
| CN102110379A (en) * | 2011-02-22 | 2011-06-29 | 黄振强 | Multimedia reading matter giving readers enhanced feeling of reality |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016005948A3 (en) * | 2014-07-11 | 2016-05-26 | Idvision Limited | Augmented reality system |
| CN107403441A (en) * | 2016-05-19 | 2017-11-28 | 视辰信息科技(上海)有限公司 | The tracking and terminal device of augmented reality system |
| CN107403441B (en) * | 2016-05-19 | 2020-11-27 | 视辰信息科技(上海)有限公司 | Tracking method of augmented reality system and terminal equipment |
| CN106096540A (en) * | 2016-06-08 | 2016-11-09 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN106096540B (en) * | 2016-06-08 | 2020-07-24 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN110248127A (en) * | 2019-06-20 | 2019-09-17 | 陕西科技大学 | A kind of broadcast transmitter radio frequency unit signal amplitude modulation circuit |
| CN110248127B (en) * | 2019-06-20 | 2021-03-26 | 陕西科技大学 | Radio frequency unit signal amplitude modulation circuit of broadcast transmitter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866496B (en) | Robot positioning and mapping method and device based on depth image | |
| US11710322B2 (en) | Surveillance information generation apparatus, imaging direction estimation apparatus, surveillance information generation method, imaging direction estimation method, and program | |
| CN104715249B (en) | Object tracking methods and device | |
| US20100097483A1 (en) | Data processing apparatus and method, and recording medium | |
| CN109074657B (en) | Target tracking method and device, electronic equipment and readable storage medium | |
| CN202502670U (en) | Image processing apparatus | |
| CN109934892B (en) | Inertial navigation motion trajectory drawing method and device and computer equipment | |
| CN104937638A (en) | Systems and methods for tracking and detecting a target object | |
| KR20130056309A (en) | Text-based 3d augmented reality | |
| CN103699869A (en) | Method and device for recognizing two-dimension codes | |
| CN103164690A (en) | Method and device for utilizing motion tendency to track augmented reality three-dimensional multi-mark | |
| TW201341758A (en) | Positioning method and system for real navigation and computer readable storage medium | |
| EP2877981A1 (en) | Method and apparatus for detecting occlusion in an augmented reality display | |
| CN109697725A (en) | A kind of filtering background method, apparatus and computer readable storage medium | |
| CN104915393A (en) | Directional search method and device | |
| CN113743177A (en) | Key point detection method, system, intelligent terminal and storage medium | |
| US20120213285A1 (en) | Motion vector detection apparatus and method | |
| CN106851584A (en) | Recognize the method and device of mobile device local environment | |
| CN101483771B (en) | Method and apparatus for promoting frame rate | |
| CN107872671B (en) | Picture coding method and terminal | |
| EP3146502B1 (en) | Accelerated image processing | |
| US9811889B2 (en) | Method, apparatus and computer program product for generating unobstructed object views | |
| CN101634557A (en) | Method for preventing picture display characters from being mutually overlapped in navigation system | |
| CN110222056A (en) | A kind of localization method, system and equipment | |
| CN109033924A (en) | The method and device of humanoid detection in a kind of video |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130619 |
|
| WD01 | Invention patent application deemed withdrawn after publication |