Multi-Scale Invertible Neural Network for Wide-Range Variable-Rate Learned Image Compression
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916664455462912 |
|---|---|
| author | Tu, Hanyue Wu, Siqi Li, Li Zhou, Wengang Li, Houqiang |
| author_facet | Tu, Hanyue Wu, Siqi Li, Li Zhou, Wengang Li, Houqiang |
| contents | Autoencoder-based structures have dominated recent learned image compression methods. However, the inherent information loss associated with autoencoders limits their rate-distortion performance at high bit rates and restricts their flexibility of rate adaptation. In this paper, we present a variable-rate image compression model based on invertible transform to overcome these limitations. Specifically, we design a lightweight multi-scale invertible neural network, which bijectively maps the input image into multi-scale latent representations. To improve the compression efficiency, a multi-scale spatial-channel context model with extended gain units is devised to estimate the entropy of the latent representation from high to low levels. Experimental results demonstrate that the proposed method achieves state-of-the-art performance compared to existing variable-rate methods, and remains competitive with recent multi-model approaches. Notably, our method is the first learned image compression solution that outperforms VVC across a very wide range of bit rates using a single model, especially at high bit rates. The source code is available at https://github.com/hytu99/MSINN-VRLIC. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_21284 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-Scale Invertible Neural Network for Wide-Range Variable-Rate Learned Image Compression Tu, Hanyue Wu, Siqi Li, Li Zhou, Wengang Li, Houqiang Computer Vision and Pattern Recognition Artificial Intelligence Autoencoder-based structures have dominated recent learned image compression methods. However, the inherent information loss associated with autoencoders limits their rate-distortion performance at high bit rates and restricts their flexibility of rate adaptation. In this paper, we present a variable-rate image compression model based on invertible transform to overcome these limitations. Specifically, we design a lightweight multi-scale invertible neural network, which bijectively maps the input image into multi-scale latent representations. To improve the compression efficiency, a multi-scale spatial-channel context model with extended gain units is devised to estimate the entropy of the latent representation from high to low levels. Experimental results demonstrate that the proposed method achieves state-of-the-art performance compared to existing variable-rate methods, and remains competitive with recent multi-model approaches. Notably, our method is the first learned image compression solution that outperforms VVC across a very wide range of bit rates using a single model, especially at high bit rates. The source code is available at https://github.com/hytu99/MSINN-VRLIC. |
| title | Multi-Scale Invertible Neural Network for Wide-Range Variable-Rate Learned Image Compression |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.21284 |