Research map / Label-efficient and robust learning / OOD and anomaly detection
PixOOD: Pixel-Level Out-of-Distribution Detection
Tomas Vojir, Jan Sochman, Jiri Matas
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 |
|---|---|---|---|
| use multi-component class modelling with soft-to-hard condensation instead of a single mean (ablation) | PixOOD · Road anomaly (RA+FS+RO mean) | mean AP: 88.14 -> 94.80 | Improved |
| use soft-to-hard condensation instead of soft K-Means (K=200) (ablation) | PixOOD · Road anomaly (RA+FS+RO mean) | mean AP: 93.01 -> 94.11 | Improved |
| increase condensation budget K from 200 to 1000 (ablation) | PixOOD · Road anomaly (RA+FS+RO mean) | mean AP: 94.11 -> 94.80 | Improved |
| use an MLP classifier instead of a linear probe (ablation) | PixOOD · Road anomaly (RA+FS+RO mean) | mean AP: 92.36 -> 94.80 | Improved |
| up-sample feature map by 7 instead of 1 (ablation) | PixOOD · Road anomaly (RA+FS+RO mean) | mean AP: 92.14 -> 94.80 | Improved |
| require patch label percentage above 90% instead of 50% (ablation) | PixOOD · Road anomaly (RA+FS+RO mean) | mean AP: 94.22 -> 94.80 | Improved |
| add DaCUP reconstruction module as a second OOD head (25% weight) (ablation) | PixOOD · Road Anomaly | AP: 96.39 -> 98.66 | Improved |
| add DRAEM as specialised module to PixOOD (ablation) | PixOOD · MVTec AD | AUPRO: 91.3 -> 95.3 | Improved |
| PixOOD without OOD training data versus methods with OOD data on SMIYC Anomaly Track | PixOOD · SMIYC Anomaly Track | AP: 68.88 vs 94.46 (RbA with OOD data) | Hurt |
| use PixOOD generic unified model on LaRS vs top leaderboard | PixOOD · LaRS | Q: 73.9 vs 78.1 | Mixed |
Closest papers by meaning
- EAT: Towards Long-Tailed Out-of-Distribution Detection (AAAI 2024 · similarity 0.85)
- MultiOOD: Scaling Out-of-Distribution Detection for Multiple Modalities (NeurIPS 2024 · similarity 0.85)
- Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection (NeurIPS 2024 · similarity 0.85)
- Rethinking the Evaluation of Out-of-Distribution Detection: A Sorites Paradox (NeurIPS 2024 · similarity 0.84)
- Out-Of-Distribution Detection with Diversification (Provably) (NeurIPS 2024 · similarity 0.84)
- Long-Tailed Out-of-Distribution Detection: Prioritizing Attention to Tail (AAAI 2025 · 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.
