Research map / Detection and segmentation / Open-set and few-shot detection
PIDray: A Large-Scale X-ray Benchmark for Real-World Prohibited Item Detection
Libo Zhang, Lutao Jiang, Ruyi Ji, Heng Fan
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 images without prohibited items to training set (ablation) | Cascade Mask R-CNN · PIDray detection | detection AP: 63.2 -> 63.7 | Improved |
| add images without prohibited items to training set (ablation) | Cascade Mask R-CNN · PIDray detection | error rate: 21.4 -> 4.8 | Improved |
| add images without prohibited items to training set (ablation) | Q2L CvT-w24-384 · PIDray multi-label classification | mAP: 90.04 -> 93.49 | Improved |
| add images without prohibited items to training set (ablation) | Q2L CvT-w24-384 · PIDray multi-label classification | error rate: 61.70 -> 6.3x10^-3 | Improved |
| add divide-and-conquer coarse-grained node (ablation) | DDOD · PIDray detection | detection AP: 63.6 -> 64.5 | Improved |
| add divide-and-conquer coarse-grained node (ablation) | Cascade Mask R-CNN · PIDray detection | detection AP: 63.7 -> 65.4 | Improved |
| add divide-and-conquer coarse-grained node (ablation) | Cascade Mask R-CNN · PIDray segmentation | segmentation AP: 52.5 -> 52.7 | No clear effect |
| add channel attention to dense attention module (ablation) | pipeline Cascade Mask R-CNN · PIDray detection | detection AP: 65.4 -> 65.7 | Improved |
| add spatial attention to dense attention module (ablation) | pipeline Cascade Mask R-CNN · PIDray detection | detection AP: 65.4 -> 65.9 | Improved |
| add dependency refinement to CA+SA (ablation) | pipeline Cascade Mask R-CNN · PIDray detection | detection AP: 66.1 -> 66.6 | Improved |
| replace GAP with class-specific cross-attention queries (ablation) | CvT-21-384 classifier · PIDray multi-label classification | mAP: 92.43 -> 92.79 | Improved |
| add divide-and-conquer pipeline to multi-label model (ablation) | Cross-Attention Module · PIDray multi-label classification | mAP: 92.79 -> 93.16 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 5 of 5 on most shared results.
- TOOD — TOOD: Task-Aligned One-Stage Object Detection (ICCV 2021) · better on 1 of 1 shared result
- Libra R-CNN — Libra R-CNN: Towards Balanced Learning for Object Detection (CVPR 2019) · better on 1 of 1 shared result
- FCOS — FCOS: Fully Convolutional One-Stage Object Detection (ICCV 2019) · better on 1 of 1 shared result
- Cascade R-CNN — Cascade R-CNN: Delving Into High Quality Object Detection (CVPR 2018) · better on 1 of 1 shared result
- Faster R-CNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 1 of 1 shared result
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
- CSPCL: Category Semantic Prior Contrastive Learning for Deformable DETR-Based Prohibited Item Detectors (NeurIPS 2025 · similarity 0.84)
- Dual-view X-ray Detection: Can AI Detect Prohibited Items from Dual-view X-ray Images like Humans? (CVPR 2025 · similarity 0.83)
- ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization (NeurIPS 2025 · similarity 0.82)
- STING-BEE: Towards Vision-Language Model for Real-World X-ray Baggage Security Inspection (CVPR 2025 · similarity 0.82)
- PiD: Generalized AI-Generated Images Detection with Pixelwise Decomposition Residuals (ICML 2025 · similarity 0.81)
- Weak-to-Strong 3D Object Detection with X-Ray Distillation (CVPR 2024 · similarity 0.81)
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.
