Research map / Detection and segmentation / CNN features
Exploring Reliable Matching with Phase Enhancement for Night-time Semantic Segmentation
Yuwen Pan, Rui Sun, Naisong Luo, Tianzhu Zhang, Yongdong Zhang
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 Fourier phase enhancement to the model with reliable matching (ablation) | NightFormer · NightCity | mIoU: 61.1 -> 63.5 | Improved |
| add reliable matching to the model with phase enhancement (ablation) | NightFormer · NightCity | mIoU: 60.4 -> 63.5 | Improved |
| add both phase enhancement and reliable matching (ablation) | NightFormer · BDD100k-night | mIoU: 49.6 -> 55.4 | Improved |
| fuse multi-level features F5,F4,F3,F2 instead of only F5 (ablation) | hierarchical amplified decoder · NightCity | mIoU: 61.5 -> 63.5 | Improved |
| use Fourier phase instead of Sobel operator for texture enhancement (ablation) | NightFormer · NightCity | mIoU: 62.1 -> 63.5 | Improved |
| add Canny operator texture enhancement instead of none (ablation) | NightFormer · NightCity | mIoU: 61.1 -> 61.9 | Improved |
| set number of prototypes N away from 32 (ablation) | NightFormer · NightCity | mIoU | Hurt |
| use too many reliable points K beyond 40 (ablation) | NightFormer · NightCity | mIoU | Hurt |
| train on NightCity plus CityScapes with NightFormer | NightFormer (Swin-Base) · CityScapes | mIoU: 78.3 -> 82.1 | Improved |
Closest papers by meaning
- The Parables of the Mustard Seed and the Yeast: Extremely Low-Budget, High-Performance Nighttime Semantic Segmentation (AAAI 2025 · similarity 0.88)
- NightAdapter: Learning a Frequency Adapter for Generalizable Night-time Scene Segmentation (CVPR 2025 · similarity 0.87)
- See through the Dark: Learning Illumination-affined Representations for Nighttime Occupancy Prediction (NeurIPS 2025 · similarity 0.84)
- NightCC: Nighttime Color Constancy via Adaptive Channel Masking (CVPR 2024 · similarity 0.84)
- Exploring the Common Appearance-Boundary Adaptation for Nighttime Optical Flow (ICLR 2024 · similarity 0.83)
- Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Driving (CVPR 2024 · similarity 0.83)
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.
