Learning Accurate Entropy Model with Global Reference for Image Compression

Yichen Qian, Zhiyu Tan, Xiuyu Sun, Ming Lin, Dongyang Li, Zhenhong Sun, Li Hao, Rong Jin

Keywords: Image compression, Entropy Model, Global Reference

Abstract: In recent deep image compression neural networks, the entropy model plays a critical role in estimating the prior distribution of deep image encodings. Existing methods combine hyperprior with local context in the entropy estimation function. This greatly limits their performance due to the absence of a global vision. In this work, we propose a novel Global Reference Model for image compression to effectively leverage both the local and the global context information, leading to an enhanced compression rate. The proposed method scans decoded latents and then finds the most relevant latent to assist the distribution estimating of the current latent. A by-product of this work is the innovation of a mean-shifting GDN module that further improves the performance. Experimental results demonstrate that the proposed model outperforms the rate-distortion performance of most of the state-of-the-art methods in the industry.

One-sentence Summary: In this paper, we propose a novel Reference-based Model for image compression to effectively leverage both the local and global context information, which yields an enhanced compression performance.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Code: [ damo-cv/img-comp-reference](https://github.com/damo-cv/img-comp-reference) + [ 1 community implementation](https://paperswithcode.com/paper/?openreview=cTbIjyrUVwJ)