TempRetinex: Retinex-based Unsupervised Enhancement for Low-light Video Under Diverse Lighting Conditions
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908929433272320 |
|---|---|
| author | Li, Yini Forster, Louis Bull, David Anantrasirichai, Nantheera |
| author_facet | Li, Yini Forster, Louis Bull, David Anantrasirichai, Nantheera |
| contents | The acquisition of paired low-light video sequences remains challenging due to issues associated with poor temporal consistency, varying illumination characteristics and camera parameters. This has driven significant interest in unsupervised low-light enhancement approaches. In this context, we propose TempRetinex, an unsupervised Retinex-based video enhancement framework exploiting inter-frame correlations. We introduce Brightness Consistency Preprocessing (BCP) that explicitly aligns intensity distributions across exposures. BCP is shown to significantly improve model robustness to diverse lighting scenarios. Moreover, we propose a multiscale temporal consistency-aware loss and an occlusion-aware masking technique to enforce similarity between consecutive frames. We further incorporate a Reverse Inference (RI) strategy to refine temporally unstable frames and a Self-Ensemble (SE) mechanism to boost denoising across diverse textures. Experiments demonstrate that TempRetinex achieves state-of-the-art performance in perceptual quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_09609 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TempRetinex: Retinex-based Unsupervised Enhancement for Low-light Video Under Diverse Lighting Conditions Li, Yini Forster, Louis Bull, David Anantrasirichai, Nantheera Image and Video Processing The acquisition of paired low-light video sequences remains challenging due to issues associated with poor temporal consistency, varying illumination characteristics and camera parameters. This has driven significant interest in unsupervised low-light enhancement approaches. In this context, we propose TempRetinex, an unsupervised Retinex-based video enhancement framework exploiting inter-frame correlations. We introduce Brightness Consistency Preprocessing (BCP) that explicitly aligns intensity distributions across exposures. BCP is shown to significantly improve model robustness to diverse lighting scenarios. Moreover, we propose a multiscale temporal consistency-aware loss and an occlusion-aware masking technique to enforce similarity between consecutive frames. We further incorporate a Reverse Inference (RI) strategy to refine temporally unstable frames and a Self-Ensemble (SE) mechanism to boost denoising across diverse textures. Experiments demonstrate that TempRetinex achieves state-of-the-art performance in perceptual quality. |
| title | TempRetinex: Retinex-based Unsupervised Enhancement for Low-light Video Under Diverse Lighting Conditions |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2511.09609 |