120 dog breeds form the base of this image classification dataset. It is an amplified version of the classic Stanford Dogs Dataset, with thousands of new high-quality images fetched from Bing and filtered via YOLO object detection. The dataset is hosted by author fedehorl on Hugging Face in Parquet format.
Use Cases
- Train image classification models based on the 120 dog breed categories.
- Benchmark fine-grained visual recognition algorithms based on breed-specific images.
- Augment training data for computer vision models based on the injected high-quality images.
- Test object detection model performance based on YOLO-filtered image content.
Strengths
- Builds upon the original Stanford Dogs Dataset with 120 established dog breed categories.
- Includes thousands of new high-quality images sourced from Bing.
- Images are filtered via YOLO object detection, likely improving data quality.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- Stanford Dogs Dataset, Bing image search.
- Collection Method
- Images were automatically fetched, cleaned, and injected from Bing, filtered dynamically via YOLO object detection.
- Freshness
- Last updated 2026-03-17 23:12:28; freshness should be verified.