ICON
ml-neuman
ICON | ml-neuman | |
---|---|---|
6 | 2 | |
1,542 | 1,240 | |
- | 0.2% | |
4.1 | 0.0 | |
5 months ago | 11 months ago | |
Python | Python | |
GNU General Public License v3.0 or later | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
ICON
- ControlNet fully integrated with Blender using nodes!
- Is there any AI that can compile several pictures of a person into a single, 3d version?
-
[R][P] ICON: Implicit Clothed humans Obtained from Normals + Gradio Web Demo
github: https://github.com/YuliangXiu/ICON
- Show HN: Icon-3D Avatar Creator from 2D Pixels
-
Icon: Towards Large-Scale Avatar Creation from In-the-Wild Pixels
Realistic virtual humans will play a central role in mixed and augmented reality, forming a critical foundation for the Metaverse and supporting remote presence, collaboration, education, and entertainment.
To enable this, new tools are needed to easily create large-scale 3D virtual humans that can be readily animated. However, current methods need either posed 3D scans captured by expensive scanning equipment or 2D images with carefully controlled user poses. Both of them can't scale up easily.
ICON ("Implicit Clothed humans Obtained from Normals") takes a step towards robust 3D clothed human reconstruction from in-the-wild images. This also enables creating animatable avatars directly from video with personalized and natural pose-dependent cloth deformation.
Homepage: https://icon.is.tue.mpg.de/
Github: https://github.com/YuliangXiu/ICON
Google Colab: https://colab.research.google.com/drive/1-AWeWhPvCTBX0KfMtgt...
ml-neuman
-
Is there any AI that can compile several pictures of a person into a single, 3d version?
for NeRF specifcally there's https://github.com/apple/ml-neuman (although the quality seems a little off to me).
-
Apple researchers propose a novel framework to reconstruct the human and the scene that can be rendered with novel human poses and views from just a single in-the-wild video
Code: https://github.com/apple/ml-neuman
What are some alternatives?
lightweight-human-pose-estimation.pytorch - Fast and accurate human pose estimation in PyTorch. Contains implementation of "Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose" paper.
rome - Realistic mesh-based avatars. ECCV 2022
ECON - [CVPR'23, Highlight] ECON: Explicit Clothed humans Optimized via Normal integration
dream-textures - Stable Diffusion built-in to Blender
AlphaPose - Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System
text2cinemagraph - Text2Cinemagraph: Text-Guided Synthesis of Eulerian Cinemagraphs [SIGGRAPH ASIA 2023]
aistplusplus_api - API to support AIST++ Dataset: https://google.github.io/aistplusplus_dataset
VIBE - Official implementation of CVPR2020 paper "VIBE: Video Inference for Human Body Pose and Shape Estimation"
Text2Video - ICASSP 2022: "Text2Video: text-driven talking-head video synthesis with phonetic dictionary".
agi2nerf - Simple tool for converting Agisoft XML files to NERF JSON files for https://github.com/NVlabs/instant-ngp