Reproduce “DoF-Query Transformer for Adaptive Fusion of Egocentric Vision and Forearm HD-sEMG in Continuous Hand Kinematics” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiahang Liu, Wenbo Zhang, Junsong Wang, Zhenzhi Ying, Siqi Hou, Naohiko Sugita, Liming Shu
The authors estimate 20 continuous hand-joint angles by combining egocentric RGB images with 128-channel forearm HD-sEMG, using a separate learned query for each joint. Across 10 participants, fusion achieved R² of 0.88 ± 0.03 and MAE of 2.80° ± 0.30°, outperforming vision-only overall. Evidence is limited to participants, setup, and held-out movements; severe synthetic occlusion reduced fusion below EMG-only.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.