BadmintonGRF: A Multimodal Dataset and Benchmark for Markerless Ground Reaction Force Estimation in Badminton

BadmintonGRF is a multi-modal dataset for ground reaction force (GRF) estimation from badminton videos.This release (v1.0) provides processed impact segments, GRF signals, synchronization metadata, and LOSO splits for 10 subjects.The dataset includes pose data from multi-view videos, GRF labels, and synchronization information.Raw videos and motion capture data are available via separate access.