Research map / Detection and segmentation / CNN features
Severity-Aware Semantic Segmentation With Reinforced Wasserstein Training
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace CE training with an adaptively learned severity-aware Wasserstein loss | Deeplab driving system · CARLA | Km/Collision: 12.4 -> 20.7 | Helped |
| replace CE training with an adaptively learned severity-aware Wasserstein loss | Deeplab driving system · CARLA | Drive%: 82.2 -> 91.6 | Helped |
| replace CE training with an adaptively learned severity-aware Wasserstein loss | ENet segmenter · CARLA training condition | distance between person collisions: 12.61 -> 30.43 | Helped |
| replace CE training with an adaptively learned severity-aware Wasserstein loss | ENet segmenter · CARLA new town | distance between person collisions: 2.53 -> 7.82 | Helped |
| replace CE training with an adaptively learned severity-aware Wasserstein loss | ENet segmenter · CARLA new weather | distance between person collisions: 9.24 -> 28.25 | Helped |
| train with an adaptively learned severity-aware Wasserstein loss instead of CE | SegNet · Cityscapes | mIoU: 57.0 -> 65.9 | Helped |
| train with an adaptively learned severity-aware Wasserstein loss instead of CE | ENet · Cityscapes | mIoU: 58.3 -> 68.2 | Helped |
| use the learned ground matrix rather than IAL | SegNet · Cityscapes | Group 4 IoU: 84.1 -> 85.4 | Helped |
| use the learned ground matrix rather than IAL | ENet · Cityscapes | Group 4 IoU: 87.7 -> 90.2 | Helped |
| use the learned ground matrix rather than IAL | ENet · Cityscapes | Bus IoU: 24.1 -> 44.2 | Helped |
| use the learned ground matrix rather than IAL | FCN · CamVid | mIoU: 71.2 -> 71.5 | Helped |
| use the learned ground matrix rather than IAL | FCN · CamVid | Group 3 IoU: 96.3 -> 97.3 | Mixed |
Closest papers by meaning
- Importance-Aware Semantic Segmentation in Self-Driving with Discrete Wasserstein Training (AAAI 2020 · similarity 0.94)
- Cross-Dataset Collaborative Learning for Semantic Segmentation in Autonomous Driving (AAAI 2022 · similarity 0.88)
- Consistent-Separable Feature Representation for Semantic Segmentation (AAAI 2021 · similarity 0.86)
- CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentation (AAAI 2020 · similarity 0.86)
- UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior (NeurIPS 2024 · similarity 0.85)
- Weakly Supervised Semantic Segmentation for Driving Scenes (AAAI 2024 · similarity 0.85)
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.