Uloženo v:
| Hlavní autoři: | , , |
|---|---|
| Médium: | Preprint |
| Vydáno: |
2024
|
| Témata: | |
| On-line přístup: | https://arxiv.org/abs/2402.15744 |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| _version_ | 1866910343189495808 |
|---|---|
| author | Li, Zhiyuan Ge, Chenyang Li, Shun |
| author_facet | Li, Zhiyuan Ge, Chenyang Li, Shun |
| contents | Recently, many deep image compression methods have been proposed and achieved remarkable performance. However, these methods are dedicated to optimizing the compression performance and speed at medium and high bitrates, while research on ultra low bitrates is limited. In this work, we propose a ultra low bitrates enhanced invertible encoding network guided by traditional transformation theory, experiments show that our codec outperforms existing methods in both compression and reconstruction performance. Specifically, we introduce the Block Discrete Cosine Transformation to model the sparsity of features and employ traditional Haar transformation to improve the reconstruction performance of the model without increasing the bitstream cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_15744 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Traditional Transformation Theory Guided Model for Learned Image Compression Li, Zhiyuan Ge, Chenyang Li, Shun Image and Video Processing Computer Vision and Pattern Recognition Recently, many deep image compression methods have been proposed and achieved remarkable performance. However, these methods are dedicated to optimizing the compression performance and speed at medium and high bitrates, while research on ultra low bitrates is limited. In this work, we propose a ultra low bitrates enhanced invertible encoding network guided by traditional transformation theory, experiments show that our codec outperforms existing methods in both compression and reconstruction performance. Specifically, we introduce the Block Discrete Cosine Transformation to model the sparsity of features and employ traditional Haar transformation to improve the reconstruction performance of the model without increasing the bitstream cost. |
| title | Traditional Transformation Theory Guided Model for Learned Image Compression |
| topic | Image and Video Processing Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.15744 |