CN106851247A - Complex scene layered approach based on depth information - Google Patents
Complex scene layered approach based on depth information Download PDFInfo
- Publication number
- CN106851247A CN106851247A CN201710077000.5A CN201710077000A CN106851247A CN 106851247 A CN106851247 A CN 106851247A CN 201710077000 A CN201710077000 A CN 201710077000A CN 106851247 A CN106851247 A CN 106851247A
- Authority
- CN
- China
- Prior art keywords
- depth
- complex scene
- arrays
- empty
- point
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/128—Adjusting depth or disparity
-
- 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
- H04N13/106—Processing image signals
- H04N13/111—Transformation of image signals corresponding to virtual viewpoints, e.g. spatial image interpolation
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Processing Or Creating Images (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of complex scene layered approach based on depth information.The present invention carries out layering to depth map and causes that image repair effect is more excellent according to the depth information of complex scene.Will not be lost according to foreground information in virtual view synthesis, loss information is the particularity of background information and complex scene, therefore the present invention is layered to complex scene using depth information in order to preferably repair, depth i.e. to complex scene is marked, by the depth difference of foreground and background, a complex scene is divided into multiple simple scenarios.The present invention ideally combines complex scene reparation with simple scenario reparation, and complex scene is divided into multiple simple scenarios by the way of depth value layering, has huge help to follow-up image repair work.
Description
Technical field
The invention belongs to the multimedia technology field of image repair, and in particular to a kind of complex scene based on depth information
Layered approach.
Background technology
To indicate spectators growing day by day to 3D content interests for nearest 3DTV and 3D films burning hot.Generally, for 3D
Video with a large amount of views, it is necessary to be used for providing three dimensional depth.But substantial amounts of view how is processed in the transmission as problem.Cause
This needs a kind of method to support to generate high-quality 3 D video using the view of limited quantity.
In the past, due to the expensive price of depth cameras, so depth information is very unobtainable.But present,
The depth camera of high-performance low price becomes generally.Synthesize virtual view using color diagram depth map corresponding with its
Method is referred to as DIBR technologies.Using depth and camera information, during all of pixel projected into three-dimensional world again.Again by generation
Pixel in boundary's coordinate is projected in the image plane of virtual view again, here it is so-called virtual view synthetic technology.It is empty
Intending View Synthesis technology can reduce substantial amounts of memory space in 3 D video transmission and propagate bandwidth.
But have a key issue in virtual view synthetic technology, part background after viewpoint change is carried out, by
Moved in the relative position of prospect background, caused from the sightless background of former viewpoint in another viewpoint, so as to new
The hole region of information is produced without in the image of viewpoint, also a kind of situation is because depth map depth inaccurately, causes picture
The cavity that the relative movement distance of vegetarian refreshments changes and occurs.
For virtual view synthetic technology, because inaccurate or depth map the missing of depth map is while in scene
Hold relative complex situation, cause being not easily repaired for cavity.Therefore need to recover depth map before virtual view synthesis,
So as to ensure the correctness of virtual view synthesis.Certain forefathers also studied certain methods, but be directed to complex scene, simply
Means are repaired in filtering method or general layering can not good repairing effect.
The content of the invention
The present invention is directed to deficiency of the existing Depth Stratification technology in complex scene application aspect, according to the depth of complex scene
Information carries out layering to depth map and causes that image repair effect is more excellent.Will not be lost according to foreground information in virtual view synthesis,
Loss information is the particularity of background information and complex scene, therefore the present invention is entered using depth information to complex scene
In order to preferably be repaired, i.e., the depth to complex scene is marked for row layering, by the depth difference of foreground and background,
One complex scene is divided into multiple simple scenarios.
The technical solution adopted for the present invention to solve the technical problems is as follows：
Step one：If depth map is I, cavity point rgb pixel value is set to (0 255 0), travels through all pixels of depth map
Point, sets empty array fillregion, and the point that pixel is empty pixel value is set into 1, non-cavity in fillregion
The point of pixel value is 0, and cavity is divided into multiple cavity blocks by the eight connectivity according to cavity point.
Step 2：To each empty block, if empty edge isP set up an office at edgeOn, traversalUpper each p point,
The corresponding depth value of p points is Z (p), is that deep minimum is added to FG arrays by Z (p) min, and Z (p) max are that depth maximum adds
BG arrays are added to, FG arrays and BG arrays is obtained.
Step 3：FG arrays and BG arrays are traveled through, FGmax therein and BGmin is obtained, if FGmax>
BGmin, by the small image-region of background parts depth value, i.e., from camera closer to that empty block and its neighboring area extract
Come, be individually layered as one and processed.
Step 4：Repeat step three, until FGmax<BGmin.Using remaining hole region as at last layering
Reason；Whole complex scene is just so divided into multiple simple scenarios, final layering is completed.
Beneficial effects of the present invention：The present invention ideally combines complex scene reparation with simple scenario reparation, adopts
Complex scene is divided into multiple simple scenarios by the mode being layered with depth value, has huge side to follow-up image repair work
Help.
Specific embodiment
Embodiments of the invention will be below described further
Embodiment is comprised the concrete steps that：
Step one：If depth map is I, cavity point rgb pixel value is set to (0 255 0) (as pure green), travels through depth
The all pixels point of figure, sets empty array fillregion, by point that pixel is empty pixel value in fillregion
1 is set to, the point of non-empty pixel value is 0, cavity is divided into multiple cavity blocks by the eight connectivity according to cavity point.
Step 2：To each empty block, if empty edge isThe p that sets up an office existsOn, traversalUpper each p point, p points
Corresponding depth value is Z (p), is that deep minimum is added to FG arrays by Z (p) min, and Z (p) max are the addition of depth maximum
To BG arrays, FG arrays and BG arrays are obtained.
Step 3：FG arrays and BG arrays are traveled through, FGmax therein and BGmin, IF FGmax is obtained>
BGmin, it is meant that the foreground part depth value of the background parts depth value less than another empty block of one of cavity block.Cause
This is needed the small image-region of background parts depth value, i.e., from camera closer to that empty block and its neighboring area extract
Come, be individually layered as one and processed.
Step 4：Repeat step three, until FGmax<BGmin.Using remaining hole region as at last layering
Reason.Its depth threshold is layered for threshold=α * FG+ β * BG to each, wherein (alpha+beta=1).
Whole complex scene is just so divided into multiple simple scenarios, final layering is completed.
Claims (1)
1. the complex scene layered approach of depth information is based on, it is characterised in that the method comprises the following steps：
Step one：If depth map is, cavity point rgb pixel value be set to (0 255 0), travel through depth map all pixels point, if
Hole array is empty, the point that pixel is empty pixel value is set to 1 in empty array, the point of non-empty pixel value is 0, according to
Cavity is divided into multiple cavity blocks by the eight connectivity of cavity point；
Step 2：To each empty block, if empty edge is, the p that sets up an office is at edgeOn, traversalUpper each p point, p points
Corresponding depth value is Z (p), is that deep minimum is added to FG arrays by Z (p) min, and Z (p) max are the addition of depth maximum
To BG arrays, FG arrays and BG arrays are obtained；
Step 3：FG arrays and BG arrays are traveled through, FGmax therein and BGmin is obtained, if FGmax>BGmin, will
The small image-region of background parts depth value, i.e., from camera closer to that empty block and its neighboring area extract, as
One single layering is processed；
Step 4：Repeat step three, until FGmax<BGmin；
Using remaining hole region as last layered shaping；Whole complex scene is just so divided into multiple simple fields
Scape, completes final layering.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710077000.5A CN106851247A (en) | 2017-02-13 | 2017-02-13 | Complex scene layered approach based on depth information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710077000.5A CN106851247A (en) | 2017-02-13 | 2017-02-13 | Complex scene layered approach based on depth information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106851247A true CN106851247A (en) | 2017-06-13 |
Family
ID=59128902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710077000.5A Pending CN106851247A (en) | 2017-02-13 | 2017-02-13 | Complex scene layered approach based on depth information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106851247A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902657A (en) * | 2010-07-16 | 2010-12-01 | 浙江大学 | Method for generating virtual multi-viewpoint images based on depth image layering |
| CN103714573A (en) * | 2013-12-16 | 2014-04-09 | 华为技术有限公司 | Virtual view generating method and virtual view generating device |
| CN104837000A (en) * | 2015-04-17 | 2015-08-12 | 东南大学 | Virtual viewpoint synthesis method using contour perception |
| CN104869386A (en) * | 2015-04-09 | 2015-08-26 | 东南大学 | Virtual viewpoint synthesizing method based on layered processing |
| CN105791803A (en) * | 2016-03-16 | 2016-07-20 | 深圳创维－Rgb电子有限公司 | Display method and system capable of converting two-dimensional image into multi-viewpoint image |
-
2017
- 2017-02-13 CN CN201710077000.5A patent/CN106851247A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902657A (en) * | 2010-07-16 | 2010-12-01 | 浙江大学 | Method for generating virtual multi-viewpoint images based on depth image layering |
| CN103714573A (en) * | 2013-12-16 | 2014-04-09 | 华为技术有限公司 | Virtual view generating method and virtual view generating device |
| CN104869386A (en) * | 2015-04-09 | 2015-08-26 | 东南大学 | Virtual viewpoint synthesizing method based on layered processing |
| CN104837000A (en) * | 2015-04-17 | 2015-08-12 | 东南大学 | Virtual viewpoint synthesis method using contour perception |
| CN105791803A (en) * | 2016-03-16 | 2016-07-20 | 深圳创维－Rgb电子有限公司 | Display method and system capable of converting two-dimensional image into multi-viewpoint image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102592275B (en) | Virtual viewpoint rendering method | |
| CN101902657B (en) | Method for generating virtual multi-viewpoint images based on depth image layering | |
| CN101771893B (en) | Video frequency sequence background modeling based virtual viewpoint rendering method | |
| CN102254348B (en) | Virtual viewpoint mapping method based o adaptive disparity estimation | |
| CN102447925B (en) | Method and device for synthesizing virtual viewpoint image | |
| CN103414909B (en) | A kind of hole-filling method being applied to dimensional video virtual viewpoint synthesis | |
| CN103810685A (en) | Super resolution processing method for depth image | |
| CN104780355A (en) | Depth-based cavity repairing method in viewpoint synthesis | |
| CN102223553A (en) | Method for converting two-dimensional video into three-dimensional video automatically | |
| CN111047709B (en) | Binocular vision naked eye 3D image generation method | |
| CN102724531B (en) | A kind of two-dimensional video turns the method and system of 3 D video | |
| CN106504190A (en) | A kind of three-dimensional video-frequency generation method based on 3D convolutional neural networks | |
| CN103702099B (en) | A kind of super large visual angle integration imaging 3D display packing based on head-tracking | |
| CN102905145B (en) | Stereoscopic image system, image generation method, image adjustment device and method thereof | |
| CN103209334A (en) | Virtual viewpoint synthesis and void repairing method for 2.5D videos to multi-view (three-dimensional) 3D videos | |
| CN103761734A (en) | Binocular stereoscopic video scene fusion method for keeping time domain consistency | |
| KR101717379B1 (en) | System for postprocessing 3-dimensional image | |
| CN104869386A (en) | Virtual viewpoint synthesizing method based on layered processing | |
| CN104270624B (en) | A kind of subregional 3D video mapping method | |
| CN103761766A (en) | Three-dimensional object model texture mapping algorithm based on tone mapping and image smoothing | |
| CN103247065B (en) | A kind of bore hole 3D video generation method | |
| CN103634584B (en) | A kind of multiple views 3D image synthesizing method | |
| CN104661014B (en) | The gap filling method that space-time combines | |
| CN103945209B (en) | A kind of DIBR method based on piecemeal projection | |
| CN101695140B (en) | Object-based virtual image drawing method of three-dimensional/free viewpoint television |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170613 |
|
| WD01 | Invention patent application deemed withdrawn after publication |