sam-3d-body is a machine learning framework for 3D human mesh recovery and pose estimation. It utilizes a 3D human mesh recovery model to reconstruct full-body meshes, including the body, hands, and feet, from a single image. The project implements a specialized extension of the Segment Anything Model to guide the extraction and refinement of human body shapes. This integration allows for prompt-guided mesh recovery, where 2D masks and keypoints constrain the inference of 3D pose and shape parameters. The system covers a range of computer vision capabilities, including 3D spatial alignment t
vrn is a 3D face reconstruction tool that generates three-dimensional volumetric representations of human faces from single two-dimensional images. It utilizes a volumetric convolutional neural network regression model to predict 3D volume data directly from image pixels. The system converts these volumetric predictions into 3D meshes through isosurface extraction and vertex coloring. It further applies realistic surface details by mapping two-dimensional image pixels onto the resulting 3D mesh using nearest-neighbor texture projection. The project provides capabilities for single-image dept
Liangxiao Hu 1,† , Hongwen Zhang 2 , Yuxiang Zhang 3 , Boyao Zhou 3 , Boning Liu 3 , Shengping Zhang 1,* , Liqiang Nie 1 ,
First dataset of dressed humans with specific geometry representation for the clothes. It contains ~2 Million images with 40 male/40 female performing 70 actions. Every subject-action sequence is captured from 4 camera views and annotated with: RGB, 3D skeleton, body part and cloth segmentation…
This repository is a reference implementation for NeuMan. NeuMan reconstructs both the background scene and an animatable human from a single video using neural radiance fields.
The main features of apple/ml-neuman are: Human Digitalization.
Open-source alternatives to apple/ml-neuman include: facebookresearch/sam-3d-body — sam-3d-body is a machine learning framework for 3D human mesh recovery and pose estimation. It utilizes a 3D human… aaronjackson/vrn — vrn is a 3D face reconstruction tool that generates three-dimensional volumetric representations of human faces from… aipixel/gaussianavatar — Liangxiao Hu 1,† , Hongwen Zhang 2 , Yuxiang Zhang 3 , Boyao Zhou 3 , Boning Liu 3 , Shengping Zhang 1,* ,… aljazbozic/deepdeform — [CVPR'2020] DeepDeform: Learning Non-rigid RGB-D Reconstruction with Semi-supervised Data. bharat-b7/ipnet — Repo for "Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction, ECCV'20 (Oral)". albertpumarola/3dpeople-dataset — First dataset of dressed humans with specific geometry representation for the clothes. It contains ~2 Million images…