Research map / Detection and segmentation / Open-set and few-shot detection
Guided Curriculum Model Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation
Christos Sakaridis, Dengxin Dai, Luc Van Gool
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 guided segmentation refinement (ablation) | GCMA · Dark Zurich-test | mIoU: 39.4% -> 42.0% | Improved |
| use GCMA curriculum with real unlabeled data instead of direct CycleGAN adaptation (ablation) | RefineNet · Dark Zurich-test | mIoU: 37.1% -> 39.4% | Improved |
| add direct CycleGAN nighttime-stylized adaptation (ablation) | RefineNet · Dark Zurich-test | mIoU: 28.5% -> 37.1% | Improved |
| apply full GCMA to daytime baseline | RefineNet · Dark Zurich-test | mIoU: 28.5% -> 42.0% | Improved |
| compare GCMA with DMAda | DMAda · Dark Zurich-test | mIoU: 32.1 -> 42.0 | Improved |
| compare GCMA with DMAda | DMAda · Nighttime Driving | mIoU: 36.1 -> 45.6 | Improved |
| adapt AdaptSegNet from Cityscapes to Dark Zurich-night | AdaptSegNet · Dark Zurich-test | mIoU: 28.8 -> 30.4 | Improved |
| adapt AdaptSegNet from Cityscapes to Dark Zurich-night | AdaptSegNet · Nighttime Driving | mIoU: 32.6 -> 34.5 | Improved |
| apply GCMA instead of DMAda on sky class | RefineNet · Dark Zurich-test | sky IoU: 0.4 -> 32.1 | Improved |
| apply GCMA instead of DMAda on motorcycle class | RefineNet · Dark Zurich-test | motorcycle IoU: 34.9 -> 30.7 | Hurt |
| use UIoU confidence thresholding instead of standard IoU | GCMA, DMAda, RefineNet · Dark Zurich-test | mean UIoU vs IoU: ca. 3% | Improved |
Closest papers by meaning
- Map-Guided Curriculum Domain Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation (TPAMI 2020 · similarity 0.98)
- The Parables of the Mustard Seed and the Yeast: Extremely Low-Budget, High-Performance Nighttime Semantic Segmentation (AAAI 2025 · similarity 0.93)
- NightAdapter: Learning a Frequency Adapter for Generalizable Night-time Scene Segmentation (CVPR 2025 · similarity 0.88)
- NightCC: Nighttime Color Constancy via Adaptive Channel Masking (CVPR 2024 · similarity 0.86)
- See through the Dark: Learning Illumination-affined Representations for Nighttime Occupancy Prediction (NeurIPS 2025 · similarity 0.86)
- Exploring the Common Appearance-Boundary Adaptation for Nighttime Optical Flow (ICLR 2024 · similarity 0.86)
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.
