COCO-2014-VL-Enriched: COCO 2014 Images with Label Issues

An enriched version of the COCO 2014 dataset. It includes label issues to help curate a cleaner dataset. The dataset was uploaded by visual-layer and last updated on September 16, 2024.

Use Cases

• Identify and correct annotation errors in object detection datasets based on the provided label issues.
• Train or evaluate data cleaning and label quality assessment algorithms using the enriched label metadata.
• Curate a higher-quality subset of the COCO 2014 dataset for model training based on the identified problematic annotations.
• Benchmark the performance of object detection models on data with known annotation inconsistencies.

Strengths

• Based on the widely-used COCO 2014 benchmark dataset for object detection.
• Enriched with label issue annotations intended to facilitate data curation.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

visual-layer

Collection Method

An enriched version of the original COCO 2014 dataset.

Freshness

Last updated 2024-09-16 12:08:06; freshness should be verified.