Estimating human pose and shape from monocular images is a long-standing problem in
computer vision. Since the release of statistical body models, 3D human mesh recovery has …

Recovering a 3D human mesh from a single RGB image is a challenging task due to depth
ambiguity and self-occlusion, resulting in a high degree of uncertainty. Meanwhile, diffusion …

As it is hard to calibrate single-view RGB images in the wild, existing 3D human mesh
reconstruction (3DHMR) methods either use a constant large focal length or estimate one …

Despite significant progress in single image-based 3D human mesh recovery, accurately
and smoothly recovering 3D human motion from a video remains challenging. Existing video …

Depth estimation is usually ill-posed and ambiguous for monocular camera-based 3D multi-
person pose estimation. Since LiDAR can capture accurate depth information in long-range …

Abstract 3D human pose estimation and mesh recovery have attracted widespread research
interest in many areas, such as computer vision, autonomous driving, and robotics. Deep …

Human mesh reconstruction (HMR) from monocular video is the key step to many mixed
reality and robotic applications. Although existing methods show promising results by …

While 3D body reconstruction methods have made remarkable progress recently, it remains
difficult to acquire the sufficiently accurate and numerous 3D supervisions required for …

Achieving the optimal balance of minimizing bandwidth consumption and end-to-end
latency while preserving a satisfactory level of visual quality becomes the ultimate goal of …

This paper presents an inverse kinematic optimization layer (IKOL) for 3D human pose and
shape estimation that leverages the strength of both optimization-and regression-based …