GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrieval
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911958152773632 |
|---|---|
| author | Zhou, Han Dong, Wei Liu, Xiaohong Liu, Shuaicheng Min, Xiongkuo Zhai, Guangtao Chen, Jun |
| author_facet | Zhou, Han Dong, Wei Liu, Xiaohong Liu, Shuaicheng Min, Xiongkuo Zhai, Guangtao Chen, Jun |
| contents | Most existing Low-light Image Enhancement (LLIE) methods either directly map Low-Light (LL) to Normal-Light (NL) images or use semantic or illumination maps as guides. However, the ill-posed nature of LLIE and the difficulty of semantic retrieval from impaired inputs limit these methods, especially in extremely low-light conditions. To address this issue, we present a new LLIE network via Generative LAtent feature based codebook REtrieval (GLARE), in which the codebook prior is derived from undegraded NL images using a Vector Quantization (VQ) strategy. More importantly, we develop a generative Invertible Latent Normalizing Flow (I-LNF) module to align the LL feature distribution to NL latent representations, guaranteeing the correct code retrieval in the codebook. In addition, a novel Adaptive Feature Transformation (AFT) module, featuring an adjustable function for users and comprising an Adaptive Mix-up Block (AMB) along with a dual-decoder architecture, is devised to further enhance fidelity while preserving the realistic details provided by codebook prior. Extensive experiments confirm the superior performance of GLARE on various benchmark datasets and real-world data. Its effectiveness as a preprocessing tool in low-light object detection tasks further validates GLARE for high-level vision applications. Code is released at https://github.com/LowLevelAI/GLARE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_12431 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrieval Zhou, Han Dong, Wei Liu, Xiaohong Liu, Shuaicheng Min, Xiongkuo Zhai, Guangtao Chen, Jun Computer Vision and Pattern Recognition Most existing Low-light Image Enhancement (LLIE) methods either directly map Low-Light (LL) to Normal-Light (NL) images or use semantic or illumination maps as guides. However, the ill-posed nature of LLIE and the difficulty of semantic retrieval from impaired inputs limit these methods, especially in extremely low-light conditions. To address this issue, we present a new LLIE network via Generative LAtent feature based codebook REtrieval (GLARE), in which the codebook prior is derived from undegraded NL images using a Vector Quantization (VQ) strategy. More importantly, we develop a generative Invertible Latent Normalizing Flow (I-LNF) module to align the LL feature distribution to NL latent representations, guaranteeing the correct code retrieval in the codebook. In addition, a novel Adaptive Feature Transformation (AFT) module, featuring an adjustable function for users and comprising an Adaptive Mix-up Block (AMB) along with a dual-decoder architecture, is devised to further enhance fidelity while preserving the realistic details provided by codebook prior. Extensive experiments confirm the superior performance of GLARE on various benchmark datasets and real-world data. Its effectiveness as a preprocessing tool in low-light object detection tasks further validates GLARE for high-level vision applications. Code is released at https://github.com/LowLevelAI/GLARE. |
| title | GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrieval |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.12431 |