Presentation

<
FLASHand: Feed-forward reLightable and Animatable Single-view Hand Reconstruction
SessionHumans & Hands - Reconstruction
DescriptionWe propose FLASHand, the first feed-forward model to reconstruct high-fidelity, relightable, and animatable 3D hand avatars from a single RGB image. By leveraging NIMBLE priors and mesh-based 2D Gaussian Splatting, FLASHand achieves instant, personalized reconstruction with disentangled appearance, supporting real-time animation and relighting.
Authors
Ling-Xiao Zhang
Institute of Computing Technology, Chinese Academy of Sciences
University of Chinese Academy of Sciences
Lin Gao
Institute of Computing Technology, Chinese Academy of Sciences
University of Chinese Academy of Sciences
Wei-Hong He
South China University of Technology
Yu-Xuan Yang
East China Normal University
Yunbing Xing
Institute of Computing Technology, Chinese Academy of Sciences
Yu-Kun Lai
Cardiff University
Yiqiang Chen
Institute of Computing Technology, Chinese Academy of Sciences
Event Type
Technical Paper
TimeMonday, 20 July 20264:05pm - 4:15pm 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
Registration Categories
Full Conference Supporter
Full Conference
Next PresentationNext PresentationAGILE: Hand-object Interaction Reconstruction from Video via Agentic Generation
Similar Presentations
Mix3R: Mixing Feed-forward Reconstruction and Generative 3D Priors for Joint Multi-view Aligned 3D Reconstruction and Pose Estimation
VFAvatar: Feed-Forward 3D Avatar Reconstruction from Casual Image Collections
EgoRelight: Egocentric Human Capture and Illumination Recovery for Relightable and Photoreal Avatar Rendering