Research map / Vision-language and multimodal / Multimodal LLMs
Light-VQA+: A Video Quality Assessment Model for Exposure Correction with Vision-Language Guidance
Xunchu Zhou, Xiaohong Liu, Yunlong Dong, Tengchuan Kou, Yixuan Gao, Zicheng Zhang, Chunyi Li, Haoning Wu, Guangtao Zhai
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 handcrafted brightness and noise features (ablation) | Light-VQA+ variant (MLP, no HVS weights) · VEC-QA | SRCC: 0.8614 -> 0.8712 | Improved |
| replace handcrafted features with CLIP local brightness/noise and Lv4 consistency features (ablation) | Light-VQA+ variant (MLP, no HVS weights) · VEC-QA | SRCC: 0.8712 -> 0.9010 | Improved |
| add CLIP global features instead of local (ablation) | Light-VQA+ variant (MLP, no HVS weights) · VEC-QA | SRCC: 0.9010 -> 0.9021 | No clear effect |
| add cross-attention fusion instead of MLP (ablation) | Light-VQA+ · VEC-QA | SRCC: 0.9010 -> 0.9177 | Improved |
| add HVS-aligned clip weights instead of simple average (ablation) | Light-VQA+ · VEC-QA | SRCC: 0.9056 -> 0.9177 | Improved |
| add HVS weights (ablation) | Light-VQA+ · OEVR-QA | SRCC: 0.7215 -> 0.7462 | Improved |
| add cross-attention fusion (ablation) | Light-VQA+ · LLVE-QA | SRCC: 0.9317 -> 0.9428 | Improved |
| use Light-VQA+ instead of Light-VQA | VQA model · OEVR-QA | SRCC: 0.5991 -> 0.7407 | Improved |
| use Light-VQA+ instead of Light-VQA | VQA model · VEC-QA | SRCC: 0.8712 -> 0.9121 | Improved |
| use Light-VQA+ instead of Light-VQA on unseen datasets | VQA model · Mixed-B cross-dataset | SRCC: 0.6358 -> 0.7328 | Improved |
| fine-tune FEC-Net with Light-VQA+ in the loss | FEC-Net · over-exposed videos | Light-VQA+ score: 69.6 -> 74.2 | Improved |
| fine-tune FEC-Net with Light-VQA+ in the loss | FEC-Net · low-light videos | Light-VQA+ score: 59.8 -> 64.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Q-Align — Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levels (ICML 2024) · better on 3 of 3 shared results
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
- Benchmarking Large Language Models for Video Quality Assessment: Structured Perceptual Prompting and Refinement (IJCV 2026 · similarity 0.84)
- Q-CLIP: Unleashing the Power of Vision-Language Models for Video Quality Assessment through Unified Cross-Modal Adaptation (ICML 2026 · similarity 0.84)
- KVQ: Boosting Video Quality Assessment via Saliency-guided Local Perception (CVPR 2025 · similarity 0.83)
- Low-Light Image Enhancement via Generative Perceptual Priors (AAAI 2025 · similarity 0.82)
- OSMamba: Omnidirectional Spectral Mamba with Dual-Domain Prior Generator for Exposure Correction (CVPR 2025 · similarity 0.81)
- UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancement (CVPR 2026 · similarity 0.81)
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.
