CORAL: A Benchmark for Structure-aware and Brain-wide Neuron Reconstruction

CORAL is a benchmark dataset for neuron reconstruction in light microscopy. It contains data for two settings: block-level reconstruction of local neuron morphology from small 3D image blocks and brain-wide reconstruction of complete neurons across whole-mouse-brain 2D slices. The dataset was created by yangzekang2000 and was last updated on 2026-05-10.

Use Cases

• Develop algorithms for local neuron morphology reconstruction based on small 3D image blocks.
• Train models for brain-wide neuron reconstruction across whole-mouse-brain 2D slices.
• Benchmark performance of neuron segmentation and tracing tools against a standard dataset.

Strengths

• Provides data for two distinct and relevant neuron reconstruction settings.
• Serves as a benchmark, which suggests a standardized evaluation framework.

Limitations

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

Provenance

Source

yangzekang2000

Freshness

Last updated 2026-05-10 16:05:29; freshness should be verified.