STAR: A Structure and Texture Aware Retinex Model

By Jun Xu, Yingkun Hou, Dongwei Ren, Li Liu, Fan Zhu, Mengyang Yu, Haoqian Wang, Ling Shao. IEEE Transactions on Image Processing, 2019

297 citations recorded by the source corpus.

Retinex theory is developed mainly to decompose an image into the illumination and reflectance components by analyzing local image derivatives. In this theory, larger derivatives are attributed to the changes in reflectance, while smaller derivatives are emerged in the smooth illumination. In this paper, we utilize exponentiated local derivatives (with an exponent <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula>) of an observed image to generate its structure map and texture map. The structure map is produced by been amplified with <inline-formula> <tex-math notation="LaTeX">$\gamma >1$ </tex-math></inline-formula>, while the texture map is generated by been shrank with <inline-formula> <tex-math notation="LaTeX">$\gamma < 1$ </tex-math></inline-formula>. To this end, we design exponential filters for the local derivatives, and present their capability on extracting accurate structure and texture maps, influenced by the choices of exponents <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula>. The extracted structure and texture maps are employed to regularize the illumination and reflectance components in Retinex decomposition. A novel Structure and Texture Aware Retinex (STAR) model is further proposed for illumination and reflectance decomposition of a single image. We solve the STAR model by an alternating optimization algorithm. Each sub-problem is transformed into a vectorized least squares regression, with closed-form solutions. Comprehensive experiments on commonly tested datasets demonstrate that, the proposed STAR model produce better quantitative and qualitative performance than previous competing methods, on illumination and reflectance decomposition, low-light image enhancement, and color correction. The code is publicly available at <uri>https://github.com/csjunxu/STAR</uri>.

Subfields: Atomic and Molecular Physics, and Optics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/76ecfcf3560ff96665154fb788c131557cde34a2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atomic and Molecular Physics, and Optics

1. Wenhan Yang
2. Jiaying Liu
3. Haofeng Huang
4. Yue Huang
5. Xinghao Ding
6. Xiaojie Guo

In Computer Graphics and Computer-Aided Design

1. Dejia Xu
2. Wenhan Yang
3. Yue Huang
4. Xinghao Ding
5. Jiaying Liu

In Computer Vision and Pattern Recognition

1. Wenhan Yang
2. Jiaying Liu
3. Haofeng Huang
4. Yue Huang
5. Xinghao Ding
6. Kai-Kuang Ma
7. Yan Yang
8. Xiaotong Tu

Citations in the corpus

• Learning a Simple Low-Light Image Enhancer from Paired Low-Light Instances (2023, methodology): “Various methods have been proposed to handle this problem [6, 17, 29, 34].”
• Toward Fast, Flexible, and Robust Low-Light Image Enhancement (2022, unknown): “As for low-light image enhancement, we compared our SCI with four recently-proposed model-based methods (including LECARM [17], SDD [10], STAR [26]), four advanced supervised learning meth-ods (including RetinexNet [3], KinD [34], FIDE [27], DRBN [30]), and four unsupervised learning methods…”
• Benchmarking Low-Light Image Enhancement and Beyond (2021, methodology): “There are also methods focusing on exploiting more effective priors (Fu et al. 2016; Guo et al. 2017; Fu et al. 2016; Cai et al. 2017; Xu and Jung 2017; Xu et al. 2019) to regularize the enhancement of illumination and reflectance layers.”