OpenStripe: Amur Tiger Biometric Features for Individual Clustering

28 Amur tiger individuals are represented by 100 processed biometric samples derived from infrared camera video segments. The dataset contains anonymized feature vectors and reconstructed stripe patterns, supporting findings from the associated OpenStripe framework paper. It was authored by Yincalos Peng and last updated on May 5, 2026.

Use Cases

• Individual tiger clustering based on reconstructed stripe skeleton features.
• Testing open-set recognition algorithms using the provided 64-dimensional feature vectors.
• Evaluating background removal and pattern reconstruction techniques on the binary texture maps.

Strengths

• Contains ground truth labels for 100 samples across 28 individual tigers.
• Provides processed biometric features (64-dimensional vectors) and reconstructed binary stripe patterns, mitigating security risks associated with raw imagery.
• Dataset size is 25.2 MB, facilitating quick download and experimentation.

Limitations

• Row count for the primary feature matrix is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset does not contain raw images or exact geospatial coordinates, limiting environmental or locational analysis.

Provenance

Source

figshare, author Yincalos Peng

Collection Method

Processed from infrared camera (PTZ) video segments collected in Northeast China Tiger and Leopard National Park.

Freshness

Last updated 2026-05-05 17:23:01; freshness should be verified.

Geography

Northeast China Tiger and Leopard National Park (coordinates masked for security).