Research map / Detection and segmentation / Open-set and few-shot detection
Beyond Pixel Uncertainty: Bounding the OoD Objects in Road Scenes
Huachao Zhu, Zelong Liu, Zhichao Sun, Yuda Zou, Gui-Song Xia, Yongchao Xu
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 |
|---|---|---|---|
| add DetSeg-R object-box refinement | RbA · FS Static | FPR95: 41.83 -> 4.38 | Improved |
| add DetSeg-R | PEBAL · Road Anomaly | FPR95: 44.58 -> 8.7 | Improved |
| add DetSeg-R | Mask2Anomaly · Road Anomaly | AP: 79.70 -> 87.57 | Improved |
| add DetSeg-R | UNO · FS L&F | FPR95: 1.30 -> 1.95 | Hurt |
| add DetSeg-R | UNO · SMIYC Anomaly Track | FPR95: 1.98 -> 10.56 | Hurt |
| add DetSeg-R | UNO · SMIYC Anomaly Track | mean F1: 62.61 -> 73.09 | Improved |
| add DetSeg-R | RbA · SMIYC Anomaly (validation IoU) | IoU: 78.08 -> 78.08 | No clear effect |
| replace S2M with DetSeg-S for threshold-free masks | S2M · FS Static | IoU: 69.99 -> 83.34 | Improved |
| replace S2M with DetSeg-S | S2M · SMIYC Anomaly (validation) | IoU: 77.54 -> 75.82 | Hurt |
| add VPHM temporal smoothing on top of DetSeg-R | Mask2Anomaly · LostAndFound test | AP: 82.95 -> 85.78 | Improved |
| add VPHM on top of DetSeg-R | UNO · LostAndFound test | AP: 80.79 -> 80.56 | No clear effect |
| add VPHM momentum smoothing | DetSeg-R · clip 02 Hanns Klemm Str 44 objectness | std of objectness (%): 4.56 -> 0.82 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Mask2Anomaly — Mask2Anomaly: Mask Transformer for Universal Open-Set Segmentation (TPAMI 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Diffusion for Out-of-Distribution Detection on Road Scenes and Beyond (ECCV 2024 · similarity 0.88)
- RUNA: Object-Level Out-of-Distribution Detection via Regional Uncertainty Alignment of Multimodal Representations (AAAI 2025 · similarity 0.85)
- Extremely Simple Multimodal Outlier Synthesis for Out-of-Distribution Detection and Segmentation (NeurIPS 2025 · similarity 0.85)
- Spotting the Unexpected (STU): A 3D LiDAR Dataset for Anomaly Segmentation in Autonomous Driving (CVPR 2025 · similarity 0.85)
- COUNTS: Benchmarking Object Detectors and Multimodal Large Language Models under Distribution Shifts (CVPR 2025 · similarity 0.84)
- AdaNeg: Adaptive Negative Proxy Guided OOD Detection with Vision-Language Models (NeurIPS 2024 · 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.
