Research map / Video / Video segmentation and tracking
Event-Guided Consistent Video Enhancement with Modality-Adaptive Diffusion Pipeline
Kanghao Chen, Zixin Zhang, Guoqiang Liang, Lutao Jiang, Zeyu Wang, Ying-Cong Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes an event-guided diffusion pipeline for low-light video enhancement that adapts to varying illumination and to missing sensing modalities (RGB or event streams).
Problem
Low-light video enhancement (LLVE) methods using RGB and event cameras are tuned for steady low-light scenes and assume all sensing modalities are always available, so they degrade under varying illumination and when a modality is lost.
Approach
A modality-adaptive diffusion pipeline guided by event data that keeps enhanced video temporally consistent and remains usable when one sensing modality (RGB or event) is unavailable; details beyond the abstract are not stated here.
Closest papers by meaning
- AnyMod-LLVE: Low-Light Video Enhancement with Modality-Agnostic Inference (ICML 2026 · similarity 0.91)
- Low-Light Video Enhancement with Synthetic Event Guidance (AAAI 2023 · similarity 0.90)
- Event-guided Unified Framework for Low-light Video Enhancement, Frame Interpolation, and Deblurring (ICCV 2025 · similarity 0.90)
- EvDiff: High Quality Video with an Event Camera (ECCV 2026 · similarity 0.89)
- EvLight++: Low-Light Video Enhancement With an Event Camera: A Large-Scale Real-World Dataset, Novel Method, and More (TPAMI 2025 · similarity 0.89)
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 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.
