superZ678/MG-HAD
The official repo for the technical report "Multi-Granularity Hand Action Detection"
This project helps you automatically identify and precisely locate fine-grained hand actions within video footage, especially in kitchen environments. It takes raw video clips as input and outputs bounding boxes around hands along with detailed action categories (e.g., 'chopping carrots' vs. 'holding knife'). This is useful for researchers in areas like human-computer interaction, robotics, or video surveillance who need to analyze complex human activities.
Use this if you need to detect and categorize intricate hand movements in videos, where existing tools only identify broad actions or whole-body movements.
Not ideal if you're looking for whole-body action detection or if your video analysis doesn't require precise hand-level detail.
Stars
10
Forks
—
Language
—
License
MIT
Category
Last pushed
Dec 02, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/superZ678/MG-HAD"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
OxWearables/stepcount
Improved Step Counting via Foundation Models for Wrist-Worn Accelerometers
OxWearables/actinet
An activity classification model based on self-supervised learning for wrist-worn accelerometer data.
guillaume-chevalier/LSTM-Human-Activity-Recognition
Human Activity Recognition example using TensorFlow on smartphone sensors dataset and an LSTM...
felixchenfy/Realtime-Action-Recognition
Apply ML to the skeletons from OpenPose; 9 actions; multiple people. (WARNING: I'm sorry that...
aqibsaeed/Human-Activity-Recognition-using-CNN
Convolutional Neural Network for Human Activity Recognition in Tensorflow