Research map / Detection and segmentation / CNN features
Watching it in Dark: A Target-aware Representation Learning Framework for High-Level Vision Tasks in Low Illumination
Yunan Li, Yihao Zhang, Shoude Li, Long Tian, DOU QUAN, Chaoneng Li, Qiguang Miao
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 two-level bi-directional alignment and target-aware masked contrastive learning | ResNet-18 · CODaN | accuracy (%): 53.32 -> 70.32 | Improved |
| add image-level alignment (ablation) | ResNet-18 · CODaN | accuracy (%): 53.32 -> 59.08 | Improved |
| add feature-level MMD alignment on top of image-level (ablation) | ResNet-18 · CODaN | accuracy (%): 59.08 -> 61.64 | Improved |
| add target-aware masking (ablation) | ResNet-18 · CODaN | accuracy (%): 61.64 -> 64.40 | Improved |
| add contrastive learning on top of masking (ablation) | ResNet-18 · CODaN | accuracy (%): 64.40 -> 70.32 | Improved |
| use bi-directional alignment instead of dark-to-bright adaptation (ablation) | ResNet-18 · CODaN | accuracy (%): 57.96 -> 61.64 | Improved |
| use bright-to-dark instead of dark-to-bright adaptation (ablation) | ResNet-18 · CODaN | accuracy (%): 57.96 -> 51.52 | Hurt |
| apply low-light enhancement before detection | YOLOv3 · ExDark detection | mAP@0.5: 58.7 -> 51.4 (EnlightenGAN) | Hurt |
| apply RUAS enhancement before classification | ResNet-50 · ExDark classification | accuracy (%): 43.15 -> 42.18 | Hurt |
| apply WiiD vs Similarity Min-Max on action recognition | I3D · ARID | accuracy (%): 52.01% -> 58.08% | Improved |
Closest papers by meaning
- FeatEnHancer: Enhancing Hierarchical Features for Object Detection and Beyond Under Low-Light Vision (ICCV 2023 · similarity 0.88)
- Unsupervised Illumination Adaptation for Low-Light Vision (TPAMI 2024 · similarity 0.87)
- You Only Look Around: Learning Illumination-Invariant Feature for Low-light Object Detection (NeurIPS 2024 · similarity 0.86)
- Motion Prior Knowledge Learning with Homogeneous Language Descriptions for Moving Infrared Small Target Detection (AAAI 2025 · similarity 0.85)
- LookWhere? Efficient Visual Recognition by Learning Where to Look and What to See from Self-Supervision (NeurIPS 2025 · similarity 0.84)
- Trash to Treasure: Low-Light Object Detection via Decomposition-and-Aggregation (AAAI 2024 · similarity 0.84)
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.
