Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ji, Yuyang, Shen, Yixuan, Zhu, Shengjie, Kong, Yu, Liu, Feng
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.26938
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908917799321600
author Ji, Yuyang
Shen, Yixuan
Zhu, Shengjie
Kong, Yu
Liu, Feng
author_facet Ji, Yuyang
Shen, Yixuan
Zhu, Shengjie
Kong, Yu
Liu, Feng
contents We present BioCoach, a biomechanics-grounded vision--language framework for fitness coaching from streaming video. BioCoach fuses visual appearance and 3D skeletal kinematics, through a novel three-stage pipeline: an exercise-specific degree-of-freedom selector that focuses analysis on salient joints; a structured biomechanical context that pairs individualized morphometrics with cycle and constraint analysis; and a vision--biomechanics conditioned feedback module that applies cross-attention to generate precise, actionable text. Using parameter-efficient training that freezes the vision and language backbones, BioCoach yields transparent, personalized reasoning rather than pattern matching. To enable learning and fair evaluation, we augment QEVD-fit-coach with biomechanics-oriented feedback to create QEVD-bio-fit-coach, and we introduce a biomechanics-aware LLM judge metric. BioCoach delivers clear gains on QEVD-bio-fit-coach across lexical and judgment metrics while maintaining temporal triggering; on the original QEVD-fit-coach, it improves text quality and correctness with near-parity timing, demonstrating that explicit kinematics and constraints are key to accurate, phase-aware coaching.
format Preprint
id arxiv_https___arxiv_org_abs_2603_26938
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From 3D Pose to Prose: Biomechanics-Grounded Vision--Language Coaching
Ji, Yuyang
Shen, Yixuan
Zhu, Shengjie
Kong, Yu
Liu, Feng
Computer Vision and Pattern Recognition
We present BioCoach, a biomechanics-grounded vision--language framework for fitness coaching from streaming video. BioCoach fuses visual appearance and 3D skeletal kinematics, through a novel three-stage pipeline: an exercise-specific degree-of-freedom selector that focuses analysis on salient joints; a structured biomechanical context that pairs individualized morphometrics with cycle and constraint analysis; and a vision--biomechanics conditioned feedback module that applies cross-attention to generate precise, actionable text. Using parameter-efficient training that freezes the vision and language backbones, BioCoach yields transparent, personalized reasoning rather than pattern matching. To enable learning and fair evaluation, we augment QEVD-fit-coach with biomechanics-oriented feedback to create QEVD-bio-fit-coach, and we introduce a biomechanics-aware LLM judge metric. BioCoach delivers clear gains on QEVD-bio-fit-coach across lexical and judgment metrics while maintaining temporal triggering; on the original QEVD-fit-coach, it improves text quality and correctness with near-parity timing, demonstrating that explicit kinematics and constraints are key to accurate, phase-aware coaching.
title From 3D Pose to Prose: Biomechanics-Grounded Vision--Language Coaching
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.26938