Instance segmentation data for Chinese character stroke extraction, featuring benchmarks for character analysis. It includes annotated images where individual strokes are labeled as separate instances to facilitate granular character decomposition.
Use Cases
- Train instance segmentation models to isolate individual strokes using the provided mask annotations
- Develop stroke-order analysis tools by extracting and sequencing individual stroke instances
- Benchmark character recognition models on their ability to decompose characters into constituent strokes
Strengths
- Features instance-level segmentation masks for individual Chinese character strokes
- Includes benchmark data for evaluating stroke extraction performance
- Focuses on character decomposition through pixel-level stroke labeling