Presentation
TwinPose: Person-Specific Subspaces for Multi-View 3D Pose Estimation
DescriptionTwinPose is an observation-driven framework for real-time multi-person 3D motion capture from sparse multi-view inputs in complex scenes. By constructing instance-aware "twin poses" that unify 2D pose semantics and multi-view geometric consistency, it enables robust, efficient, and detector-agnostic 3D human pose estimation, even under occlusions and dense interactions.
Event Type
Technical Paper
TimeMonday, 20 July 20263:45pm - 3:55pm PDT
LocationRoom 403 A
Digital Library
PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Monday, 20 July 20263:45pm - 5:35pm PDTRoom 403 A
Full Conference Supporter
Full Conference
