Search papers, labs, and topics across Lattice.
3
0
4
0
SSMB sets a new standard in keypoint detection under motion blur, achieving superior performance without relying on deblurring or handcrafted features.
Current markerless motion capture models crumble when faced with the messy reality of human interaction: occlusions, rapid movements, and similar clothing styles.
Unlike existing dynamic 3D Gaussian Splatting methods that struggle with rapid, non-rigid motions, this approach delivers temporally coherent novel views of sports and performances by revisiting neural volume rendering.
