GABIC: Graph-based Attention Block for Image Compression
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916422837338112 |
|---|---|
| author | Spadaro, Gabriele Presta, Alberto Tartaglione, Enzo Giraldo, Jhony H. Grangetto, Marco Fiandrotti, Attilio |
| author_facet | Spadaro, Gabriele Presta, Alberto Tartaglione, Enzo Giraldo, Jhony H. Grangetto, Marco Fiandrotti, Attilio |
| contents | While standardized codecs like JPEG and HEVC-intra represent the industry standard in image compression, neural Learned Image Compression (LIC) codecs represent a promising alternative. In detail, integrating attention mechanisms from Vision Transformers into LIC models has shown improved compression efficiency. However, extra efficiency often comes at the cost of aggregating redundant features. This work proposes a Graph-based Attention Block for Image Compression (GABIC), a method to reduce feature redundancy based on a k-Nearest Neighbors enhanced attention mechanism. Our experiments show that GABIC outperforms comparable methods, particularly at high bit rates, enhancing compression performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_02981 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GABIC: Graph-based Attention Block for Image Compression Spadaro, Gabriele Presta, Alberto Tartaglione, Enzo Giraldo, Jhony H. Grangetto, Marco Fiandrotti, Attilio Image and Video Processing Computer Vision and Pattern Recognition Machine Learning While standardized codecs like JPEG and HEVC-intra represent the industry standard in image compression, neural Learned Image Compression (LIC) codecs represent a promising alternative. In detail, integrating attention mechanisms from Vision Transformers into LIC models has shown improved compression efficiency. However, extra efficiency often comes at the cost of aggregating redundant features. This work proposes a Graph-based Attention Block for Image Compression (GABIC), a method to reduce feature redundancy based on a k-Nearest Neighbors enhanced attention mechanism. Our experiments show that GABIC outperforms comparable methods, particularly at high bit rates, enhancing compression performance. |
| title | GABIC: Graph-based Attention Block for Image Compression |
| topic | Image and Video Processing Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2410.02981 |