Research map / Video / Video segmentation and tracking
RetinexMCNet: A Memory Controller Dominated Network for Low-Light Video Enhancement Based on Retinex
Meiao Wang, Xuejing Kang, Yaxi Lu, Jie Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add memory controller (MC) to IAM+RDM (ablation) | RetinexMCNet · SDSD | MSE(MABD): 0.502 -> 0.237 | Improved |
| add Illumination Normalization (IN) (ablation) | RetinexMCNet · SDSD | PSNR: 27.63 -> 27.71 | Improved |
| add LTS loss (ablation) | RetinexMCNet · SDSD | PSNR: 27.71 -> 27.74 | Improved |
| use all components (IN and LTS together) vs MC alone (ablation) | RetinexMCNet · SDSD | PSNR: 27.63 -> 27.81 | Improved |
| use IAM and RDM instead of naive UNet for illumination (ablation) | RetinexMCNet · SDSD | PSNR: 25.40 -> 27.81 | Improved |
| use RDM instead of naive Transformer (ablation) | RetinexMCNet · SDSD | PSNR: 27.64 -> 27.81 | Improved |
| make RetinexFormer MC-dominated | RetinexFormer (LLIE) · Consistency | MSE(MABD): 0.545 -> 0.342 | Improved |
| make BRVE MC-dominated | BRVE (adjacent frames) · Consistency | MSE(MABD): 0.511 -> 0.337 | Improved |
| make StableLLVE MC-dominated | StableLLVE (random frames) · Consistency | MSE(MABD): 0.623 -> 0.366 | Improved |
| use historical keyframes instead of adjacent frames as S (ablation) | IAM+RDM · SDSD | PSNR: 27.33 -> 27.81 | Improved |
| use historical keyframes instead of random frames as S (ablation) | IAM+RDM · SDSD | PSNR: 27.38 -> 27.81 | Improved |
| use historical keyframes instead of adjacent frames as S (ablation) | RetinexFormer · SDSD | PSNR: 26.65 -> 27.37 | Improved |
Closest papers by meaning
- IniRetinex: Rethinking Retinex-type Low-Light Image Enhancer via Initialization Perspective (AAAI 2025 · similarity 0.90)
- Multinex: Lightweight Low-light Image Enhancement via Multi-prior Retinex (CVPR 2026 · similarity 0.90)
- Spike-RetinexFormer: Rethinking Low-light Image Enhancement with Spiking Neural Networks (NeurIPS 2025 · similarity 0.89)
- You Do Not Need Additional Priors or Regularizers in Retinex-Based Low-Light Image Enhancement (CVPR 2023 · similarity 0.89)
- Unrolled Decomposed Unpaired Learning for Controllable Low-Light Video Enhancement (ECCV 2024 · similarity 0.89)
- MambaLLIE: Implicit Retinex-Aware Low Light Enhancement with Global-then-Local State Space (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
