Research map / Image generation / Image restoration
Multinex: Lightweight Low-light Image Enhancement via Multi-prior Retinex
Alexandru Brateanu, Tingting Mu, Codruta O. Ancuti, Cosmin Ancuti
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 luminance guidance stack and luminance term to raw RGB network (ablation) | Multinex · LOLv1 | PSNR: 14.15 -> 20.57 | Improved |
| add reflectance term to luminance-only model (ablation) | Multinex · LOLv1 | PSNR: 20.57 -> 23.19 | Improved |
| add luminance term to reflectance-only model (ablation) | Multinex · LOLv1 | PSNR: 18.50 -> 23.19 | Improved |
| add CWA to model without CWA or MSEF (ablation) | Multinex fusion module · LOLv1 | PSNR: 19.92 -> 21.34 | Improved |
| add MSEF to model without CWA or MSEF (ablation) | Multinex fusion module · LOLv1 | PSNR: 19.92 -> 22.27 | Improved |
| add CWA to MSEF-only fusion module (ablation) | Multinex fusion module · LOLv1 | PSNR: 22.27 -> 23.19 | Improved |
| use CWA instead of CBAM spatial attention (ablation) | Multinex · LOLv1 | PSNR: 20.44 -> 23.19 | Improved |
| use CWA instead of MHSA with pooling (ablation) | Multinex · LOLv1 | PSNR: 21.67 -> 23.19 | Improved |
| use CWA instead of MDTA (ablation) | Multinex · LOLv1 | PSNR: 22.39 -> 23.19 | Improved |
| use Multinex instead of LYT-Net at similar size | lightweight LLIE · LOLv1 | PSNR: 22.38 -> 23.19 | Improved |
| use Multinex instead of CIDNet | lightweight LLIE · LOLv1 | PSNR: 23.81 -> 23.19 | Hurt |
| use Multinex-Nano instead of RSFNet | micro LLIE · LOLv2-syn | PSNR: 17.18 -> 21.05 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.
- RetinexFormer — Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement (ICCV 2023) · better on 1 of 3 shared results
- GLARE — GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrieval (ECCV 2024) · better on 0 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- IniRetinex: Rethinking Retinex-type Low-Light Image Enhancer via Initialization Perspective (AAAI 2025 · similarity 0.91)
- You Do Not Need Additional Priors or Regularizers in Retinex-Based Low-Light Image Enhancement (CVPR 2023 · similarity 0.91)
- Nonlocal Retinex-Based Variational Model and its Deep Unfolding Twin for Low-Light Image Enhancement (IJCV 2025 · similarity 0.90)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.89)
- Enhancing Low-Light Images: A Synthetic Data Perspective on Practical and Generalizable Solutions (AAAI 2025 · similarity 0.88)
- MambaLLIE: Implicit Retinex-Aware Low Light Enhancement with Global-then-Local State Space (NeurIPS 2024 · similarity 0.88)
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.
