Fig. 6From: Hands-on: deformable pose and motion models for spatiotemporal localization of fine-grained dyadic interactionsExample frames from ShakeFive2, SBU Kinect, and UT-Interaction. Top row: handshake, bottom row: hugBack to article page