Monocular, One-stage, Regression of Multiple 3D People and their 3D positions & trajectories in camera & global coordinates. ROMP[ICCV21], BEV[CVPR22], TRACE[CVPR2023]

Measure the SMPL body model

[CVPR 2023] Code repository for HuManiFlow: Ancestor-Conditioned Normalising Flows on SO(3) Manifolds for Human Pose and Shape Distribution Estimation

➿A rerun plugin and tools for 3D animation

[ECCV 24 wksp] Pose-independent 3D Anthropometry from Sparse Data

The Fast Way From Vertices to Parametric 3D Humans

Compress a stream of nonparametric 3D human mesh estimates

Like SMPL but for bugs

This project generates a 3D dancing mesh figure for a song by analyzing beat, tempo, and emotion to produce accurate dance steps. It also designs costumes and background settings based on the music's emotional tone, enhancing user experience. Additionally, we created a Bharatnatyam dance dataset using OpenPose, Mediapipe, and SMPL points.

A simple and fast GUI tool to visualize and compare the SMPL sequences and scenes in real-time.

[ICCV 2021, Oral] PyMAF: 3D Human Pose and Shape Regression with Pyramidal Mesh Alignment Feedback Loop

Dockerfile for mmhuman3d.

This project was born out of a passion for computer vision and 3D human modeling. It explores the power of SMPLX for body shape representation and demonstrates its application in generating realistic 3D avatars from simple measurements.

Turning humans into 汉字 (Hànzì) characters.

A set of tools to visualize and interact with sequences of 3D data.

Fit an SMPL body model to a scan

[T-PAMI 2022] Out-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation

Smplify-X implementation. (2024. 08. 30 No Error & Recent version)

Mesh2SMPL is a project that utilizes the MultiviewSMPLifyX and PaMIR projects to convert a textured mesh scan of a human into a SMPL model.

GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models (CVPR 2024)