ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Coding

By Dailan He, Zi Yang, Weikun Peng, Rui Ma, Hongwei Qin, Yan Wang. Computer Vision and Pattern Recognition, 2022

581 citations recorded by the source corpus.

Recently, learned image compression techniques have achieved remarkable performance, even surpassing the best manually designed lossy image coders. They are promising to be large-scale adopted. For the sake of practicality, a thorough investigation of the architecture design of learned image compression, regarding both compression performance and running speed, is essential. In this paper, we first propose uneven channel-conditional adaptive coding, motivated by the observation of energy compaction in learned image compression. Combining the proposed uneven grouping model with existing context models, we obtain a spatial-channel contextual adaptive model to improve the coding performance without damage to running speed. Then we study the structure of the main transform and propose an efficient model, ELIC, to achieve state-of-the-art speed and compression ability. With superior performance, the proposed model also supports extremely fast preview decoding and progressive decoding, which makes the coming application of learning-based image compression more promising.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f5aea8045382a7fa87331126784b432ac3fde635

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Jiro Katto
2. Zhihao Duan
3. Justin Yang
4. Jinming Liu
5. Zhan Ma
6. F. Zhu
7. Jiangpeng He
8. Heming Sun

Citations in the corpus

• Towards Backward-Compatible Continual Learning of Image Compression (2024, background): “A major group focuses on designing expressive architectures for !enc and f ctec, such as convolutional and transformer-based ones [10, 11, 14, 20, 23, 27, 33, 37-39, 43, 52, 54, 59].”
• Learned Image Compression with Mixed Transformer-CNN Architectures (2023, methodology): “According to previous works [13, 26] and discussion1, we round and encode each dy−μ] to the bitstream instead of dyc and restore the coding-symbol ŷ as dy − μc + μ, which can further benefit entropy models.”