深度学习图像压缩研究者的福音,compressai
- bmshj2018-factorized: Ballé J, Laparra V, Simoncelli E P. End-to-end optimized image compression[C]//ICLR 2017
- bmshj2018-hyperprior: Ballé J, Minnen D, Singh S, et al. Variational image compression with a scale hyperprior[C]//ICLR2018.
- mbt2018-mean: Minnen D, Ballé J, Toderici G. Joint autoregressive and hierarchical priors for learned image compression[C]//NIPS 2018.
- mbt2018: Minnen D, Ballé J, Toderici G. Joint autoregressive and hierarchical priors for learned image compression[C]//NIPS 2018.
- cheng2020-anchor: Cheng Z, Sun H, Takeuchi M, et al. Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules[C]//CVPR 2020
- cheng2020-attn: Cheng Z, Sun H, Takeuchi M, et al. Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules[C]//CVPR 2020
注意: bmshj2018-factorized代码里使用的熵编码方法是Variational image compression with a scale hyperprior提出的全分解方法。官方的tensorflow库里也改了的。
很多最新的模型都是基于compressai改的。感谢大佬。https://github.com/InterDigitalInc/CompressAI
另外,CSDN的叶大佬写的compressai的介绍,附上链接:http://t.csdn.cn/bhLL2
@article{begaint2020compressai,
title={CompressAI: a PyTorch library and evaluation platform for end-to-end compression research},
author={B{'e}gaint, Jean and Racap{'e}, Fabien and Feltman, Simon and Pushparaja, Akshay},
year={2020},
journal={arXiv preprint arXiv:2011.03029},
}