Oxford-IIIT Pet: 37 Pet Categories with 200 Images Each

A dataset of pet images across 37 categories, with roughly 200 images per class. The images exhibit large variations in scale, pose, and lighting. This instance includes standard train/test splits and was last updated on the platform on 2024-01-07.

Use Cases

• Fine-grained pet breed classification based on the 37 category labels.
• Evaluating model robustness to variations in scale, pose, and lighting mentioned in the description.
• Training image recognition models on a dataset with standard train/test splits.

Strengths

• 37 distinct pet categories.
• Approximately 200 images per class, providing balanced representation.
• Images have large variations in scale, pose, and lighting, which can aid model generalization.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Trimaps and bounding box annotations are not included in this instance, requiring separate reference.

Provenance

Source

Oxford Visual Geometry Group (VGG).

Freshness

Last updated 2024-01-07 18:10:47; freshness should be verified.