Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2604.06655 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914456134483968 |
|---|---|
| author | Ding, Ding Li, Daowen Chen, Ying Gao, Yixin Dong, Ruixiao Li, Kai Li, Li |
| author_facet | Ding, Ding Li, Daowen Chen, Ying Gao, Yixin Dong, Ruixiao Li, Kai Li, Li |
| contents | Perceptual video compression adopts generative video modeling to improve perceptual realism but frequently sacrifices signal fidelity, diverging from the goal of video compression to faithfully reproduce visual signal. To alleviate the dilemma between perception and fidelity, in this paper we propose Controllable Generative Video Compression (CGVC) paradigm to faithfully generate details guided by multiple visual conditions. Under the paradigm, representative keyframes of the scene are coded and used to provide structural priors for non-keyframe generation. Dense per-frame control prior is additionally coded to better preserve finer structure and semantics of each non-keyframe. Guided by these priors, non-keyframes are reconstructed by controllable video generation model with temporal and content consistency. Furthermore, to accurately recover color information of the video, we develop a color-distance-guided keyframe selection algorithm to adaptively choose keyframes. Experimental results show CGVC outperforms previous perceptual video compression method in terms of both signal fidelity and perceptual quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_06655 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Controllable Generative Video Compression Ding, Ding Li, Daowen Chen, Ying Gao, Yixin Dong, Ruixiao Li, Kai Li, Li Computer Vision and Pattern Recognition Perceptual video compression adopts generative video modeling to improve perceptual realism but frequently sacrifices signal fidelity, diverging from the goal of video compression to faithfully reproduce visual signal. To alleviate the dilemma between perception and fidelity, in this paper we propose Controllable Generative Video Compression (CGVC) paradigm to faithfully generate details guided by multiple visual conditions. Under the paradigm, representative keyframes of the scene are coded and used to provide structural priors for non-keyframe generation. Dense per-frame control prior is additionally coded to better preserve finer structure and semantics of each non-keyframe. Guided by these priors, non-keyframes are reconstructed by controllable video generation model with temporal and content consistency. Furthermore, to accurately recover color information of the video, we develop a color-distance-guided keyframe selection algorithm to adaptively choose keyframes. Experimental results show CGVC outperforms previous perceptual video compression method in terms of both signal fidelity and perceptual quality. |
| title | Controllable Generative Video Compression |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.06655 |