Research map / Label-efficient and robust learning / OOD and anomaly detection
A Likelihood Ratio-Based Approach to Segmenting Unknown Objects
Nazir Nayal, Youssef Shoeb, Fatma Güney
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 |
|---|---|---|---|
| use DINOv2-b instead of Swin-b backbone (ablation) | UEM · RoadAnomaly | AP: 65.42 -> 92.86 | Improved |
| use DINOv2-b instead of CLIP backbone (ablation) | UEM · RoadAnomaly | AP: 87.84 -> 92.86 | Improved |
| use DINOv2-b instead of Swin-b backbone (ablation) | inlier segmentor · Cityscapes | mIoU: 81.6 -> 82.8 | Improved |
| use DINOv2-b instead of CLIP backbone (ablation) | inlier segmentor · Cityscapes | mIoU: 77.8 -> 82.8 | Improved |
| use likelihood ratio score instead of ID density only (ablation) | UEM (DINOv2-b) · RoadAnomaly | AP: 64.77 -> 92.86 | Improved |
| use likelihood ratio score instead of OoD density only (ablation) | UEM (DINOv2-b) · RoadAnomaly | AP: 88.54 -> 92.86 | Improved |
| use likelihood ratio score instead of ID density only (ablation) | UEM (DINOv2-b) · FS LaF | AP: 15.00 -> 79.02 | Improved |
| use likelihood ratio score instead of OoD density only (ablation) | UEM (DINOv2-b) · FS LaF | AP: 63.53 -> 79.02 | Improved |
| use likelihood ratio score instead of OoD only (ablation) | UEM (Swin-b) · FS LaF | FPR: 18.10 -> 18.88 | Mixed |
| use DINOv2 backbone with PEBAL scoring | PEBAL · RoadAnomaly | AP: 92.98 | Improved |
| use DINOv2 backbone with PEBAL scoring | PEBAL · FS LaF | AP: 51.11 | Hurt |
| use DINOv2 backbone with RbA scoring | RbA · RoadAnomaly | AP: 89.37 | Mixed |
Closest papers by meaning
- Extremely Simple Multimodal Outlier Synthesis for Out-of-Distribution Detection and Segmentation (NeurIPS 2025 · similarity 0.91)
- Residual Pattern Learning for Pixel-Wise Out-of-Distribution Detection in Semantic Segmentation (ICCV 2023 · similarity 0.90)
- Entropy Maximization and Meta Classification for Out-of-Distribution Detection in Semantic Segmentation (ICCV 2021 · similarity 0.89)
- Envisioning Outlier Exposure by Large Language Models for Out-of-Distribution Detection (ICML 2024 · similarity 0.88)
- Image Background Serves as Good Proxy for Out-of-distribution Data (ICLR 2024 · similarity 0.88)
- BOOD: Boundary-based Out-Of-Distribution Data Generation (ICML 2025 · similarity 0.88)
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.
