Summary of Mamballie: Implicit Retinex-aware Low Light Enhancement with Global-then-local State Space, by Jiangwei Weng et al.
MambaLLIE: Implicit Retinex-Aware Low Light Enhancement with Global-then-Local State Space
by Jiangwei Weng, Zhiqiang Yan, Ying Tai, Jianjun Qian, Jian Yang, Jun Li
First submitted to arxiv on: 25 May 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A recent advancement in low light image enhancement is dominated by the Retinex-based learning framework, which leverages convolutional neural networks (CNNs) and Transformers. However, vanilla Retinex theory primarily addresses global illumination degradation, neglecting local issues like noise and blur in dark conditions. Furthermore, CNNs and Transformers struggle to capture global degradation due to their limited receptive fields. To address this limitation, researchers propose MambaLLIE, an implicit Retinex-aware low light enhancer featuring a global-then-local state space design. This design integrates Local-Enhanced State Space Module (LESSM) and Implicit Retinex-aware Selective Kernel module (IRSK) with LayerNorm as its core. The proposed architecture enables comprehensive global long-range modeling and flexible local feature aggregation, outperforming state-of-the-art CNN and Transformer-based methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low light image enhancement is important for improving images taken in dark conditions. Recently, a new approach called MambaLLIE has been developed to enhance these types of images. This method uses a combination of existing techniques to improve the quality of low light images. The goal of this research is to make it easier to take high-quality photos even in very low light conditions. |
Keywords
» Artificial intelligence » Cnn » Transformer
