Multi-modal Human Mesh Recovery (SMPL) using image and LiDAR data

Human 3D model partiality representation via Mean Curvature Flow

Enables users to create a profile saved only to local storage and get size recommendations across sites. This is the ultimate size recommendation plugin, saving time & money for online shoppers as it uses CV to predict body measurements and GPT 4 to provide size recommendations specific to a brand.

A 100% compatiable SMPL,SMPL-H,SMPL-X model implemention in C++ with CUDA support. Same api with python smplx.

A way to visualize clothes on custom body measurements.

"Linear Regression vs. Deep Learning". The source code for a simple but effective baseline method for human body measurement estimation using only height and weight information about the person.

Code used in the GRADE framework to convert SMPL animation data to the USD file format to be used in the IsaacSim/Omniverse simulators.

AIST++ Dataset Webpage: https://google.github.io/aistplusplus_dataset

➿A rerun plugin and tools for 3D animation

A real time virtual try-on application using SMPL models and OpenCV

The Fast Way From Vertices to Parametric 3D Humans

A Wiki on Body-Modelling Technology, maintained by Meshcapade GmbH.

Body measurements

Measure the SMPL body model

Official implementation of CVPR2020 paper "Learning to Dress 3D People in Generative Clothing" https://arxiv.org/abs/1907.13615

CVPR 2022 - Official code repository for the paper: Accurate 3D Body Shape Regression using Metric and Semantic Attributes.

API to support AIST++ Dataset: https://google.github.io/aistplusplus_dataset

[CVPR 2023] Official implementation of the paper "One-Stage 3D Whole-Body Mesh Recovery with Component Aware Transformer"

[CVPR'22] ICON: Implicit Clothed humans Obtained from Normals