Micro-OD: 252 Images for Few-Shot Cell Detection in Microscopy

Micro-OD is a benchmark of 252 images curated for in-context learning, with bounding-box annotations for 11 cell types across four sources. It was created by Shreyan Ganguly and last updated in May 2026. The dataset is designed to evaluate vision-language models for few-shot object detection in biomedical microscopy.

Use Cases

• Benchmarking few-shot object detection performance based on the 252 annotated images
• Evaluating in-context learning for vision-language models based on the described test-time reasoning tokens
• Training hybrid few-shot object detection pipelines based on the combination of a detection head and VLM classifier described
• Studying domain adaptation from natural to microscopic images based on the described domain gap

Strengths

• Contains 252 specifically curated images for in-context learning
• Includes bounding-box annotations spanning 11 distinct cell types
• Derives from four sources, including two expert-annotated in-lab sets

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• The dataset is small in scale at 211.7 KB

Provenance

Source

figshare

Collection Method

Curated collection for a research benchmark, as described in the associated paper.

Freshness

Last updated 2026-05-15 05:42:51