Stochastic Scene-Aware Motion Prediction

GLAMR: Global Occlusion-Aware Human Mesh Recovery with Dynamic Cameras

“…Human Motion Modeling. Extensive research has studied 3D human dynamics for various tasks including motion prediction and synthesis [2,4,18,19,25,35,56,61,73,75,93,[100][101][102][103]. Recent human pose estimation methods start to leverage learned human dynamics models to improve the accuracy of estimated motions [45,78,111].…”

GOAL: Generating 4D Whole-Body Motion for Hand-Object Grasping

“…More relevant are methods for generating motion between a "start" and a "goal" pose in a 3D scene. Hassan et al [17] estimate a "goal" position and interaction direction on an object, plan a 3D path from a "start" body pose to this, and finally generate a sequence of body poses with an autoregressive cVAE for walking and interacting, e.g., sitting on a chair. Wang et al [58] first estimate several "sub-goal" positions and bodies, divide these into short start/end pairs to synthesize short-term motions, and finally stitch these together in a long motion with an optimization process.…”

“…the body and the speed of motion. Therefore, to generate motion of arbitrary length, we use an autoregressive network architecture [17,54].…”

Learning Body-Aware 3D Shape Generative Models

“…Scene-Aware Body Generation: There has been recent interest in the problem of generating human poses that fit a particular 3D scenes. POSA [7] uses a body-centric representation to place it at a static within 3D scenes, while SAMPL [5] places dynamic and plausible motions within 3D scenes. Our goal is to solve the inverse problem: given human body poses, synthesize plausible 3D shapes.…”