WO2018157835A1 - 基于运动注意力模型的360度全景视频编码方法 - Google Patents
基于运动注意力模型的360度全景视频编码方法 Download PDFInfo
- Publication number
- WO2018157835A1 WO2018157835A1 PCT/CN2018/077730 CN2018077730W WO2018157835A1 WO 2018157835 A1 WO2018157835 A1 WO 2018157835A1 CN 2018077730 W CN2018077730 W CN 2018077730W WO 2018157835 A1 WO2018157835 A1 WO 2018157835A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- motion
- motion vector
- attention
- block
- reliability
- Prior art date
Links
Images
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
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/513—Processing of motion vectors
-
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- 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
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/527—Global motion vector estimation
Definitions
- Step 4 Construct a sports attention model to obtain the motion attention of the coded block
- step 1 the motion vector reliability is defined as follows:
- v x and v y are components of the motion vector v in the x and y directions, respectively, and NF is a normalization factor.
- the spatial motion vector contrast is defined as follows:
- Step 1 Extract the motion vector to obtain the motion vector field, and calculate the reliability of each motion vector.
- the motion vector reliability g(v) is defined as follows:
- FIG. 3 is a motion vector field after reliability weighted filtering; the motion vector in the five circled portions of FIG. 2 is determined to be unreliable, and the motion vector around the circle is determined to be reliable, then After step 2, the unreliable motion vector in the circle is replaced by a reliable motion vector around it. At the same time, the motion vector elsewhere in the circle is judged to be reliable, and is left unreserved.
- Step 4 Construct a motion attention model to obtain the motion attention of the coding block; after the above global motion compensation processing, construct the motion vision model according to the modified motion vector, including three aspects: motion vector strength, airspace motion Vector contrast and time domain motion vector contrast.
- the motion vector strength is defined as follows:
- MA MI + MC s + MC t.
- the whiter area indicates that the calculated motion attention is greater, such as the contour edge of the moving human body. These areas are more concerned with the human eye, and the more codewords are assigned during encoding; the darker the calculated motion The smaller the attention, such as the vast majority of the background image, the human eye is less sensitive due to being at rest, and the codewords allocated during encoding are relatively small.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Description
Claims (6)
- 一种基于运动注意力模型的360度全景视频编码方法,其特征在于包括如下步骤:步骤1:提取运动矢量得到运动矢量场,计算每个运动矢量的可靠度;步骤2:根据可靠度进行可靠度的加权滤波预处理,以降低噪声;步骤3:将步骤2修正后的运动矢量场进行全局运动补偿;步骤4:构建运动注意力模型,得到编码块的运动注意力;步骤5:根据步骤5得到编码块的运动注意力,自适应分配码字。
- 根据权利要求2所述的一种基于运动注意力模型的360度全景视频编码方法,其特征在于:步骤2中进行可靠度加权滤波:若g(v)大于0.1,表示当前块运动矢量可靠,则不用做任何处理;若g(v)小于0.1,表示当前块运动矢量不可靠,则对不可靠运动矢量做可靠度加权向量中值滤波,使其周围可靠的运动矢量代替当前不可靠运动矢量。
- 根据权利要求1所述的一种基于运动注意力模型的360度全景视频编码方法,其特征在于:步骤3中:将步骤3修正后的运动矢量场进行全局运动补偿,即统计当前帧中所有SKIP模式的运动矢量的均值,并对当前帧所有运动矢量减去该均值,得到全局运动补偿后运动矢量场。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710122773.0 | 2017-03-03 | ||
CN201710122773.0A CN107454418B (zh) | 2017-03-03 | 2017-03-03 | 基于运动注意力模型的360度全景视频编码方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018157835A1 true WO2018157835A1 (zh) | 2018-09-07 |
Family
ID=60486227
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2018/077730 WO2018157835A1 (zh) | 2017-03-03 | 2018-03-01 | 基于运动注意力模型的360度全景视频编码方法 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN107454418B (zh) |
WO (1) | WO2018157835A1 (zh) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107454418B (zh) * | 2017-03-03 | 2019-11-22 | 叠境数字科技(上海)有限公司 | 基于运动注意力模型的360度全景视频编码方法 |
CN108271020B (zh) * | 2018-04-24 | 2019-08-09 | 福州大学 | 一种基于视觉注意力模型的全景视频质量评价方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050140781A1 (en) * | 2003-12-29 | 2005-06-30 | Ming-Chieh Chi | Video coding method and apparatus thereof |
CN101282479A (zh) * | 2008-05-06 | 2008-10-08 | 武汉大学 | 基于感兴趣区域的空域分辨率可调整编解码方法 |
CN102572380A (zh) * | 2010-12-29 | 2012-07-11 | 中国移动通信集团公司 | 视频监控编码方法及其装置 |
CN103765898A (zh) * | 2011-09-02 | 2014-04-30 | 索尼公司 | 图像处理设备、图像处理方法和程序 |
CN107454418A (zh) * | 2017-03-03 | 2017-12-08 | 叠境数字科技(上海)有限公司 | 基于运动注意力模型的360度全景视频编码方法 |
-
2017
- 2017-03-03 CN CN201710122773.0A patent/CN107454418B/zh active Active
-
2018
- 2018-03-01 WO PCT/CN2018/077730 patent/WO2018157835A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050140781A1 (en) * | 2003-12-29 | 2005-06-30 | Ming-Chieh Chi | Video coding method and apparatus thereof |
CN101282479A (zh) * | 2008-05-06 | 2008-10-08 | 武汉大学 | 基于感兴趣区域的空域分辨率可调整编解码方法 |
CN102572380A (zh) * | 2010-12-29 | 2012-07-11 | 中国移动通信集团公司 | 视频监控编码方法及其装置 |
CN103765898A (zh) * | 2011-09-02 | 2014-04-30 | 索尼公司 | 图像处理设备、图像处理方法和程序 |
CN107454418A (zh) * | 2017-03-03 | 2017-12-08 | 叠境数字科技(上海)有限公司 | 基于运动注意力模型的360度全景视频编码方法 |
Also Published As
Publication number | Publication date |
---|---|
CN107454418A (zh) | 2017-12-08 |
CN107454418B (zh) | 2019-11-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10805629B2 (en) | Video compression through motion warping using learning-based motion segmentation | |
CN107439012B (zh) | 用于环内转换的方法、装置和计算机可读存储介质 | |
Xu et al. | Region-of-interest based conversational HEVC coding with hierarchical perception model of face | |
Rao et al. | A Survey of Video Enhancement Techniques. | |
US20100060783A1 (en) | Processing method and device with video temporal up-conversion | |
Li et al. | Weight-based R-λ rate control for perceptual HEVC coding on conversational videos | |
Dong et al. | Human visual system-based saliency detection for high dynamic range content | |
CN102158712B (zh) | 一种基于视觉的多视点视频信号编码方法 | |
Lee et al. | Efficient video coding based on audio-visual focus of attention | |
CN108122206A (zh) | 一种低照度图像降噪方法及装置 | |
JPH07203435A (ja) | 歪んだ図形情報の強調方法及び装置 | |
US20130279598A1 (en) | Method and Apparatus For Video Compression of Stationary Scenes | |
JP2008048243A (ja) | 画像処理装置、画像処理方法および監視カメラ | |
Yang et al. | Spatiotemporal generative adversarial network-based dynamic texture synthesis for surveillance video coding | |
Yang et al. | Learning event guided high dynamic range video reconstruction | |
WO2018157835A1 (zh) | 基于运动注意力模型的360度全景视频编码方法 | |
Wu et al. | Virtual reality video quality assessment based on 3d convolutional neural networks | |
US20060133472A1 (en) | Spatial scalable compression | |
US10049436B1 (en) | Adaptive denoising for real-time video on mobile devices | |
JP4596211B2 (ja) | 画像処理装置および方法、記録媒体、並びにプログラム | |
Kwon et al. | New video enhancement preprocessor using the region-of-interest for the videoconferencing | |
US12094089B2 (en) | Intelligent video enhancement system | |
CN112364200B (zh) | 一种类脑成像的方法、装置、设备及存储介质 | |
JP5303399B2 (ja) | 動画像ビット深度削減装置及びプログラム | |
KR20120019662A (ko) | 전송될 동영상의 전처리 방법 및 이를 이용한 동영상 전송시스템. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18760473 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18760473 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 02.01.2020) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18760473 Country of ref document: EP Kind code of ref document: A1 |