JunkyByte/easy_ViTPose
Easy and fast 2d human and animal multi pose estimation using SOTA ViTPose [Y. Xu et al., 2022] Real-time performances and multiple skeletons supported.
This project helps quickly and accurately identify the body posture (pose) of multiple humans or animals in images or videos. It takes standard image or video files as input and outputs the precise location of key body joints, known as keypoints. People working in fields like sports analytics, wildlife monitoring, behavioral research, or animation would find this useful for analyzing movement.
231 stars. No commits in the last 6 months.
Use this if you need to detect and track the body keypoints of multiple subjects (humans or animals) in real-time video feeds or large collections of images with high accuracy.
Not ideal if your primary goal is object detection or classification rather than detailed pose estimation, or if you require 3D pose reconstruction.
Stars
231
Forks
39
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Sep 10, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/JunkyByte/easy_ViTPose"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
veb-101/keras-vision
Porting vision models to Keras 3 for easily accessibility. Contains MobileViT v1, MobileViT v2, fastvit
vishal-n2403/Unsupervised-Domain-Adaptation-and-ViTs
ViT + MAE for UDA on Sentinel-1/2 (SAR/optical) land-cover classification with CORAL & DANN. PyTorch.
mach-12/vit
(ViT) Vision Image Transformer PyTorch Implementation
ceasedfonts/PicReco
Image Classification using Google ViT, Transformers and Streamlit