OpenDMC
OpenDMC is An Open-Source Algorithm Library of Deep-learning-based Multi-frame Coding (DMC). We collect methods on DMC, provide source codes of MindSpore, PyTorch or TensorFlow, and test their performances.
Contact and References
Coordinator: Asst. Prof. Wei Gao (Shenzhen Graduate School, Peking University)
Should you have any suggestions for better constructing this open source library, please contact the coordinator via Email: gaowei262@pku.edu.cn. We welcome more participants to submit your codes to this collection, and you can send your OpenI ID to the above Email address to obtain the accessibility.
List of Contributors
Contributors:
Asst. Prof. Wei Gao (Shenzhen Graduate School, Peking University)
Mr. Hua Ye (Peng Cheng Laboratory)
Mr. Yongchi Zhang (Peng Cheng Laboratory)
Mr. Yuyang Wu (Shenzhen Graduate School, Peking University)
etc.
Table of Content
1.1 DCVC (by Yuyang Wu, Hua Ye)
1.2 ssf2020 (by Yuyang Wu, Hua Ye)
1.3 DVC (by Hua Ye, Yongchi Zhang)
1.1 DCVC (by Yuyang Wu, Hua Ye)
- NeurIPS 2021
- In this paper, we propose a deep contextual video compression framework to enable a paradigm shift from predictive coding to conditional coding. In particular, we try to answer the following questions: how to define, use, and learn condition under a deep video compression framework.
- Code in the framework of pytorch & mindspore are available.
- For more information, please go to DCVC.
Figure 1: The framework of DCVC, from Ref. [Li J, Li B, Lu Y. Deep contextual video compression[J]. Advances in Neural Information Processing Systems, 2021, 34: 18114-18125.]
1.2 ssf2020 (by Yuyang Wu, Hua Ye)
- CVPR 2020
- In this paper, we show that a generalized warping operator that better handles common failure cases, e.g. disocclusions and fast motion, can provide competitive compression results with a greatly simplified model and training procedure. Specifically, we propose scale-space flow, an intuitive generalization of optical flow that adds a scale parameter to allow the network to better model uncertainty.
- Code in the framework of pytorch & mindspore are available.
- For more information, please go to ssf2020.
Figure 2: Overview of the end-to-end optimized, low-latency compression system, from Ref. [Agustsson E, Minnen D, Johnston N, et al. Scale-space flow for end-to-end optimized video compression[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020: 8503-8512.]
1.3 DVC (by Hua Ye, Yongchi Zhang)
- CVPR 2019
- This paper, taking advantage of both classical architecture in the conventional video compression method and the powerful non-linear representation ability of neural networks, proposes the first end-to-end video compression deep model that jointly optimizes all the components for video compression.
- Code in the framework of pytorch & mindspore are available.
- For more information, please go to DVC.
Figure 3: The proposed end-to-end video compression network, from Ref. [Lu G, Ouyang W, Xu D, et al. Dvc: An end-to-end deep video compression framework[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019: 11006-11015.]
1.4 DVC_P (by Shangkun Sun, Yongchi Zhang)
- VCIP 2021
- This paper is based on OpenDVC, and improves it with perceptual optimizations (i.e., a discriminator network and a mixed loss are employed to help network trade off among distortion, perception and rate, and nearest-neighbor interpolation is used to eliminate checkerboard artifacts).
- Code in the framework of tensorflow & mindspore are available.
- For more information, please go to DVC_P.
Figure 4: The proposed video compression framework of DVC_P, from Ref. [Saiping Zhang, Marta Mrak, Luis Herranz, et al. DVC-P: Deep Video Compression with Perceptual Optimizations[C]// arXiv preprint arXiv:2109.10849.]