Cross-Spectral Face Recognition
A face recognition technology that compares identities using images from different imaging bands, such as visible light and thermal infrared. Its core challenge is learning modality-invariant identity representations, and it is used for nighttime security and identity verification under unlit conditions.
Timeline (last 90 days)
Most existing cross-spectral face recognition benchmarks have a gallery size of about 100 people (N ≈ 100 subjects), which limits the credibility of evaluation.
The VTF900 dataset includes expression, pose, and occlusion variations, among which pose and occlusion are the most common sources of failure for cross-spectral recognition in real-world scenarios.
All Facts (2)
Most existing cross-spectral face recognition benchmarks have a gallery size of about 100 people (N ≈ 100 subjects), which limits the credibility of evaluation.
50%UnverifiedThe VTF900 dataset includes expression, pose, and occlusion variations, among which pose and occlusion are the most common sources of failure for cross-spectral recognition in real-world scenarios.
50%