Patent
Video image recompression method
Hao Xiaolong,Feng Min,Fan Qiang,Peng Qiwei,Wang Xueguang,Luo Wang,Han Bin,Jia Zheng,Zhang Tiexun,Cui Yang +9 more
- 16 Apr 2019
TL;DR: Wang et al. as mentioned in this paper proposed a video image recompression method by combining the existing video image compression method and depth learning technology, and the compression proportion of the intra-frame data of the video image can be greatly improved.
read more
Abstract: The invention discloses a video image recompression method. The method comprises the following steps: adopting a method for segmenting and reconstructing an original video, classifying various video clips formed by segmenting the original video according to the SBD technology, processing the video clips in different types, and omitting certain frames. The invention proposes a video image recompression method by combining the existing video image compression method and depth learning technology, and the compression proportion of the intra-frame data of the video image can be greatly improved. Redundancy information between the video image frames is greatly reduced by combining the depth learning method on the existing compression method, and more than five times of compression ratio advantage can be improved on the existing method.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
References
•Book
Video Compression
Peter D. Symes
- 01 Mar 1998
TL;DR: In this article, the authors present a practical, plain-English guide to image compression, including JPEG, MPEG-1, and MPEG-2, and present an intriguing glimpse of other systems currently in development.
319
Patent
Hierarchical hybrid shot change detection method for MPEG-compressed video
Sung Bae Jun,Kyoungro Yoon +1 more
- 06 Mar 2001
TL;DR: In this article, a hierarchical hybrid shot change detection method was proposed for MPEG-compressed video circumstances, wherein the method according to the present invention comprises: detecting a shot change candidate region, by hierarchically applying a global color distribution difference among the frames by the color histogram; detecting an abridged shot change region by using the difference of the global color histograms difference among adjacent I, P frames after additionally decoding P frames existed in the detected candidate region; decoding B frames within a shot-change detection candidate region at a macro block level in a sequence of the B
55
Patent
Method realized by computer for detecting scene boundaries in videos
Naveen Goela,Kevin W. Wilson,Ajay Divakaran,Feng Niu,Isao Otsuka +4 more
- 02 Feb 2011
TL;DR: In this paper, a computer implemented method detects scene boundaries in videos by first extracting feature vectors from videos of different genres and then classifying the feature vectors as scene boundaries using a support vector machine.
3
Patent
Detection method for HEVC prediction mode steganography
Wang Rangding,Sheng Qi,Yan Diqun,Wang Bin,Li Qian +4 more
- 26 Oct 2016
TL;DR: In this paper, a detection method for HEVC prediction mode steganography is proposed, which has the advantages that the detection effect on various steganographic algorithms of modifying prediction modes is very good; the computing complexity is relatively low; and the detecting effect on the videos with different resolutions under different quantization parameters is relatively good.
3
Patent
Rate-distortion optimization coding method based on RGB source video
Wang Xu,Li Yang,Zhang Jianye,Chen Tao,Teng Yangxin,Feng Li,Guo Qingrui,Qing Song,Li Mingxuan,Zhao Gaofeng,Fan Qiang,Peng Qiwei,Luo Wang,Feng Min,Xia Yuan +14 more
- 22 Mar 2017
TL;DR: In this article, a rate-distortion optimization coding method based on RGB source video is proposed, which is characterized in comprising the steps of S1, inputting the RGB source image, S2, converting the RGB image into a YUV video, S3, carrying out predictive coding on the YUV image obtained through conversion, and S4, calculating the coding bit number for each coding mode.
1