COMPASS-XP: Matched Photographic and X-Ray Images of Objects

Lewis D. Griffin at University College London created COMPASS-XP, a dataset of matched photographic and X-ray images of single objects. Objects are imaged in multiple poses and include metadata labels for dangerousness in aviation contexts. Object classes overlap with those in the ImageNet and WordNet datasets.

Use Cases

• Train object classification models based on matched photographic and X-ray views.
• Develop multimodal image fusion algorithms based on paired image types.
• Build threat detection systems for aviation security based on dangerousness labels.
• Benchmark computer vision models using classes shared with ImageNet and WordNet.

Strengths

• Provides paired photographic and X-ray images of the same objects.
• Includes metadata labels for dangerousness in aviation contexts.
• Object classes are aligned with popular ImageNet and WordNet datasets.

Limitations

• Row count, file formats, and column-level documentation are unknown.
• Last update date is unknown; freshness unverified.

Provenance

Source

University College London