A dataset for evaluating open-vocabulary object detectors in security X-ray scans, introduced in a paper accepted at ICCV 2025. The dataset was created by researchers including Pablo Garcia-Fernandez and Elisa Ricci. It was last updated on July 17, 2025.
Use Cases
- Benchmarking open-vocabulary object detection models based on X-ray imagery.
- Training models to detect novel objects in security screening scenarios.
- Advancing research in X-ray vision for security applications as described in the paper.
Strengths
- Dataset is associated with peer-reviewed research presented at ICCV 2025.
- The repository is the official release for the described research paper.
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
Provenance
- Source
- huggingface
- Freshness
- Last updated 2025-07-17 13:24:19; freshness should be verified.