CoolMoves: User Motion Accentuation in Virtual Reality
summary
System synthesizes expressive full-body avatar motion from only headset and hand tracking by matching user cues to a mocap repository, using a nearest-neighbor search and weighted interpolation, then temporally smoothing with joint prediction probability and human motion priors. It claims robust results even with very sparse databases (3-5 motions per action), produces context-aware foot placement and scaled limb remapping, and validates effectiveness through one technical reconstruction test plus three user studies on motion quality, embodiment and agency.
AI-written summary · about
related talks
11mMove with Style! Enhancing Avatar Embodiment in Virtual Reality through Proprioceptive Motion Fee...
20mMalleable Embodiment: Changing Sense of Embodiment by Spatial-Temporal Deformation of Virtual ...
10mNon-isomorphic Interaction Techniques for Controlling Avatar Facial Expressions in VR
14mEmoShortcuts: Emotionally Expressive Body Augmentation for Social Mixed Reality Avatars
11mCan You Move These Over There? Exploring an LLM-based VR Mover to Support Natural Multi-object Ma...
09mAvatar Tracking Control with Generations of Physically Natural Responses on Contact to Reduce ...
48mTalking VR Turkey
16mMagicalHands: Mid-Air Hand Gestures for Animating in VR