vichetkao/table_layout_dataset_v1

Table Dataset - Image & LabelMe & OBB Annotation (Train/Val Split) Dataset Overview Comprehensive table detection dataset with ground truth LabelMe polygon annotations and OBB (Oriented Bounding Box) data, split into training and validation sets. Total examples: 10,000 image-annotation pairs Train: 8,000 (80.0%) Validation: 2,000 (20.0%) Total size: 670.25 MB Language: km Document types: Table/Chart documents Ground truth: LabelMe polygon annotations… See the full description on the dataset page: https://huggingface.co/datasets/vichetkao/table_layout_dataset_v1.