MSc student Eitan Gerber attended the 2026 IEEE EMBS Body Sensor Networks (BSN) conference in Portugal and presented a poster of his paper “Hand motion estimation in the wild via IMUs and gaze-tracking: Can an efficient linear model suffice?”
The paper uses data from Meta’s ‘Project Nymeria’ dataset to explore whether a method as simple as linear regression can accurately predict the hand’s 3D position based on measurements proximal to the elbow – in other words, measurements that could come from a transhumeral amputee. Eitan further compared linear regression’s prediction performance to that of a UNet based predictor, focusing on both methods’ accuracy, training time and inference time.
At the end of the conference, Eitan also participated in the “Wearable Sensing, Adaptive Robotics, and AI for Personalized Neurorehabilitation and Assistive Health Technology” workshop, taking the opportunity to learn about cutting-edge research and experiment best practices from a number of researchers in the AERIALIST and TAILOR doctoral programs.
Related publications
E. Gerber and N. E. Krausz, “Hand motion estimation in the wild via IMUs and gaze-tracking: Can an efficient linear model suffice?” in 2026 IEEE EMBS International Conference on Body Sensor Networks (BSN), Oct. 2026
Leave a Reply