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

2026-05-03

AI Summary

This work addresses the challenge of markerless ground reaction force (GRF) estimation in non-periodic sports such as badminton, which has been hindered by the absence of publicly available multimodal datasets combining high-frame-rate, synchronized multi-view video with GRF measurements. To bridge this gap, we present and release the first rigorously aligned multimodal dataset for badminton impact scenarios, integrating 120 FPS multi-view RGB videos, force data from four Kistler force plates, and motion capture from a Vicon system. The dataset comprises 17,425 impact segments and 12,867 view instances from 10 participants. Modality synchronization is ensured through event alignment, temporal offset calibration, and a multi-view deduplication mechanism. We further provide preprocessing code, a leave-one-subject-out cross-validation split, and ten baseline models to support research on markerless GRF estimation.

Abstract

Multimodal resources for non-periodic court sports with laboratory-grade sensing remain scarce: few publicly pair instrumented ground reaction force (GRF) with high-frame-rate multi-view video, limiting markerless load estimation in realistic training settings. BadmintonGRF records eight synchronized RGB views at ~120 FPS, four Kistler force plates, and Vicon motion capture (C3D) without hardware genlock across modalities; alignment combines human-verified events, automated quality assurance, and per-camera time offsets with uncertainty metadata. Tier 1 distributes pose, time-aligned GRF, metadata, and splits under CC BY-NC 4.0, enabling the primary benchmark without raw RGB or C3D; we report a Tier 1 task that maps 2D pose to GRF. Tier 2 provides raw RGB and C3D under controlled access for studies that require appearance or full kinematics. The public release contains 17,425 impact-segment archives in the 10-subject benchmark tree (156 instrumented trials; raw multi-view RGB alone exceeds 1 TB); benchmark loader gates retain 12,867 view-specific instances and 1,732 unique impacts after multi-view deduplication. We are not aware of prior public badminton corpora that combine this sensing layout with audited video--GRF alignment for impact-centric GRF estimation. We distribute preprocessing code, leave-one-subject-out splits, ten reference baselines, and optional late fusion (one deterministic test-time pass per instance; no test-time augmentation), with a within-trial diagnostic in the supplementary material.

Problem

• ground reaction force estimation
• markerless motion analysis
• multimodal dataset
• badminton
• non-periodic sports

Innovation

• markerless GRF estimation
• multimodal dataset
• badminton biomechanics
• video--force alignment
• impact-centric benchmark