Research map / Detection and segmentation / Open-set and few-shot detection
Pixel-Wise Anomaly Detection in Complex Driving Scenes
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 |
|---|---|---|---|
| add uncertainty maps to guide the dissimilarity module (ablation) | Pixel-wise anomaly detection framework · FS L&F | AP: 39 ± 9 -> 55 ± 5 | Helped |
| add uncertainty maps to guide the dissimilarity module (ablation) | Pixel-wise anomaly detection framework · FS Static | AP: 38 ± 8 -> 62 ± 5 | Helped |
| add a training data generator that includes void-class anomalies, together with uncertainty maps (ablation) | Pixel-wise anomaly detection framework · FS L&F | AP: 15 ± 5 -> 55 ± 5 | Helped |
| add a training data generator that includes void-class anomalies, together with uncertainty maps (ablation) | Pixel-wise anomaly detection framework · FS Static | AP: 14 ± 4 -> 62 ± 5 | Helped |
| ensemble dissimilarity output with uncertainty predictions using grid-searched weights (ablation) | Pixel-wise anomaly detection framework · FS L&F | FPR95: 66 ± 8 -> 40 ± 5 | Helped |
| ensemble dissimilarity output with uncertainty predictions using grid-searched weights (ablation) | Pixel-wise anomaly detection framework · FS Static | FPR95: 41 ± 12 -> 26 ± 1 | Helped |
| ensemble dissimilarity output with uncertainty predictions using grid-searched weights (ablation) | Pixel-wise anomaly detection framework · FS L&F | AP: 58 ± 9 -> 55 ± 5 | Hurt |
| ensemble dissimilarity output with uncertainty predictions using grid-searched weights (ablation) | Pixel-wise anomaly detection framework · FS Static | AP: 57 ± 6 -> 62 ± 5 | Helped |
| replace grid-searched ensemble weights with end-to-end learned weights (ablation) | Pixel-wise anomaly detection framework · FS L&F | FPR95: 39.6 -> 58.6 | Hurt |
| replace grid-searched ensemble weights with end-to-end learned weights (ablation) | Pixel-wise anomaly detection framework · FS Static | FPR95: 25.6 -> 37.3 | Hurt |
| replace grid-searched ensemble weights with end-to-end learned weights (ablation) | Pixel-wise anomaly detection framework · FS L&F | AP: 55.1 -> 59.6 | Helped |
| replace grid-searched ensemble weights with end-to-end learned weights (ablation) | Pixel-wise anomaly detection framework · FS Static | AP: 61.5 -> 61.1 | Hurt |
Closest papers by meaning
- A Dataset for Semantic Segmentation in the Presence of Unknowns (CVPR 2025 · similarity 0.90)
- Spotting the Unexpected (STU): A 3D LiDAR Dataset for Anomaly Segmentation in Autonomous Driving (CVPR 2025 · similarity 0.90)
- Self-Supervised Likelihood Estimation with Energy Guidance for Anomaly Segmentation in Urban Scenes (AAAI 2024 · similarity 0.89)
- Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection (AAAI 2024 · similarity 0.86)
- MetaUAS: Universal Anomaly Segmentation with One-Prompt Meta-Learning (NeurIPS 2024 · similarity 0.86)
- SDAC: A Multimodal Synthetic Dataset for Anomaly and Corner Case Detection in Autonomous Driving (AAAI 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.