JPEG AI Image Compression Visual Artifacts: Detection Methods and Dataset
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929587149078528 |
|---|---|
| author | Tsereh, Daria Mirgaleev, Mark Molodetskikh, Ivan Kazantsev, Roman Vatolin, Dmitriy |
| author_facet | Tsereh, Daria Mirgaleev, Mark Molodetskikh, Ivan Kazantsev, Roman Vatolin, Dmitriy |
| contents | Learning-based image compression methods have improved in recent years and started to outperform traditional codecs. However, neural-network approaches can unexpectedly introduce visual artifacts in some images. We therefore propose methods to separately detect three types of artifacts (texture and boundary degradation, color change, and text corruption), to localize the affected regions, and to quantify the artifact strength. We consider only those regions that exhibit distortion due solely to the neural compression but that a traditional codec recovers successfully at a comparable bitrate. We employed our methods to collect artifacts for the JPEG AI verification model with respect to HM-18.0, the H.265 reference software. We processed about 350,000 unique images from the Open Images dataset using different compression-quality parameters; the result is a dataset of 46,440 artifacts validated through crowd-sourced subjective assessment. Our proposed dataset and methods are valuable for testing neural-network-based image codecs, identifying bugs in these codecs, and enhancing their performance. We make source code of the methods and the dataset publicly available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_06810 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | JPEG AI Image Compression Visual Artifacts: Detection Methods and Dataset Tsereh, Daria Mirgaleev, Mark Molodetskikh, Ivan Kazantsev, Roman Vatolin, Dmitriy Artificial Intelligence Computer Vision and Pattern Recognition Multimedia Learning-based image compression methods have improved in recent years and started to outperform traditional codecs. However, neural-network approaches can unexpectedly introduce visual artifacts in some images. We therefore propose methods to separately detect three types of artifacts (texture and boundary degradation, color change, and text corruption), to localize the affected regions, and to quantify the artifact strength. We consider only those regions that exhibit distortion due solely to the neural compression but that a traditional codec recovers successfully at a comparable bitrate. We employed our methods to collect artifacts for the JPEG AI verification model with respect to HM-18.0, the H.265 reference software. We processed about 350,000 unique images from the Open Images dataset using different compression-quality parameters; the result is a dataset of 46,440 artifacts validated through crowd-sourced subjective assessment. Our proposed dataset and methods are valuable for testing neural-network-based image codecs, identifying bugs in these codecs, and enhancing their performance. We make source code of the methods and the dataset publicly available. |
| title | JPEG AI Image Compression Visual Artifacts: Detection Methods and Dataset |
| topic | Artificial Intelligence Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2411.06810 |