Estimating Second-Person Human Poses, Shapes and Global Motion from Monocular egocentric Videos
4D Human Body Capture from Egocentric Video via 3D Scene Grounding
We introduce a novel task of reconstructing a time series of second-person 3dhuman body meshes from monocular egocentric videos.We propose a novel optimization-based approach that leverages 2d observations of the entire video sequence and human-scene interaction constraint to estimate second-personhuman poses, shapes and global motion that are grounded on the 3d environment captured from the egocentric view.We conduct detailed ablation studies to validate our design choice.We compare our method with previous state-of-the-art method on human motion capture from monocular video, and show that our method estimates more accurate human-body poses and shapes under the challenging egocentric setting.In addition, we demonstrate that our approachproduces more realistic human-scene interaction.
