SoleCoach: Sole Pressure and IMU-based MLLMs for Skill Coaching
The study aims to develop SoleCoach, a portable sports training system that estimates alpine skiers’ 3D postures and generates expert-level coaching feedback using a multimodal large language model based solely on data from wearable insole sensors.
Ground Reaction Inertial Poser: Physics-based Human Motion Capture from Sparse IMUs and Insole Pressure Sensors
GRIP reconstructs physically plausible full-body motion by fusing sparse IMU data with foot pressure from Moticon OpenGo insoles in a physics simulation.
Distinguishing Gait Patterns in PD Patients under Different Treatments via Recurrence Plots and Vision Transformer Fusion
The study aims to establish a deep learning framework using recurrence plots and Vision Transformers to objectively distinguish gait signatures in Parkinson’s patients receiving different treatments compared to healthy controls.
FootFormer: Estimating Stability from Visual Input
The study aims to develop a unified deep learning architecture called FootFormer that jointly estimates dense foot pressure, ground contact, and center of mass from video sequences to quantify human postural stability.
Step2Motion: Locomotion Reconstruction from Pressure Sensing Insoles
Foot Pressure-Based Abnormal Gait Recognition With Multi-Scale Cross-Attention Fusion
A Proposal for Automatic Evaluation of Human Functional Limitations in Activities of Daily Living
This work introduces a novel proposal designed to automate the assessment of human functional limitations.
