Adapting Learned Image Codecs to Screen Content via Adjustable Transformations
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911784948989952 |
|---|---|
| author | Dogaroglu, H. Burak Koyuncu, A. Burakhan Boev, Atanas Alshina, Elena Steinbach, Eckehard |
| author_facet | Dogaroglu, H. Burak Koyuncu, A. Burakhan Boev, Atanas Alshina, Elena Steinbach, Eckehard |
| contents | As learned image codecs (LICs) become more prevalent, their low coding efficiency for out-of-distribution data becomes a bottleneck for some applications. To improve the performance of LICs for screen content (SC) images without breaking backwards compatibility, we propose to introduce parameterized and invertible linear transformations into the coding pipeline without changing the underlying baseline codec's operation flow. We design two neural networks to act as prefilters and postfilters in our setup to increase the coding efficiency and help with the recovery from coding artifacts. Our end-to-end trained solution achieves up to 10% bitrate savings on SC compression compared to the baseline LICs while introducing only 1% extra parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_17544 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Adapting Learned Image Codecs to Screen Content via Adjustable Transformations Dogaroglu, H. Burak Koyuncu, A. Burakhan Boev, Atanas Alshina, Elena Steinbach, Eckehard Image and Video Processing Computer Vision and Pattern Recognition Machine Learning As learned image codecs (LICs) become more prevalent, their low coding efficiency for out-of-distribution data becomes a bottleneck for some applications. To improve the performance of LICs for screen content (SC) images without breaking backwards compatibility, we propose to introduce parameterized and invertible linear transformations into the coding pipeline without changing the underlying baseline codec's operation flow. We design two neural networks to act as prefilters and postfilters in our setup to increase the coding efficiency and help with the recovery from coding artifacts. Our end-to-end trained solution achieves up to 10% bitrate savings on SC compression compared to the baseline LICs while introducing only 1% extra parameters. |
| title | Adapting Learned Image Codecs to Screen Content via Adjustable Transformations |
| topic | Image and Video Processing Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2402.17544 |