Research map / Label-efficient and robust learning / OOD and anomaly detection
Diffusion for Out-of-Distribution Detection on Road Scenes and Beyond
Silvio Galesso, Philipp Schröppel, Hssan Driss, Thomas Brox
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 |
|---|---|---|---|
| replace 2D U-Net denoiser with a per-feature MLP denoiser (ablation) | DOoD · OoD detection on road scenes | OoD detection quality | Improved |
| use directional (cosine) error score instead of MSE of estimated score (ablation) | DOoD · RoadAnomaly and other benchmarks | AP | Improved |
| use score-based OoD scores instead of reconstruction-based MSE (ablation) | DOoD · RoadAnomaly and other benchmarks | AP | Improved |
| aggregate OoD scores across 25 timesteps instead of 5 (ablation) | DOoD Segmenter-B · RoadAnomaly | AP: 86.1 -> 89.1 | Improved |
| use SETR-L features rather than Segmenter-B on Fishyscapes L&F (ablation) | DOoD · Fishyscapes L&F val | AP: 43.0 -> 64.4 | Improved |
| use low-resolution transformer feature maps on small obstacles | DOoD · Fishyscapes L&F and SMIYC Obstacle | FPR@95TPR: 27.7 FPR on Fishyscapes L&F (SETR-L) | Hurt |
Closest papers by meaning
- Image Background Serves as Good Proxy for Out-of-distribution Data (ICLR 2024 · similarity 0.89)
- Rethinking the Evaluation of Out-of-Distribution Detection: A Sorites Paradox (NeurIPS 2024 · similarity 0.88)
- GOOD: Training-Free Guided Diffusion Sampling for Out-of-Distribution Detection (NeurIPS 2025 · similarity 0.88)
- ImageNet-OOD: Deciphering Modern Out-of-Distribution Detection Algorithms (ICLR 2024 · similarity 0.88)
- Extremely Simple Multimodal Outlier Synthesis for Out-of-Distribution Detection and Segmentation (NeurIPS 2025 · similarity 0.88)
- A Systematic Analysis of Out-of-Distribution Detection Under Representation and Training Paradigm Shifts (NeurIPS 2026 · 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.
