kenshohara/3D-ResNets
3D ResNets for Action Recognition
This project helps researchers and developers working with video data to train and test advanced AI models that can automatically identify actions happening within video clips. It takes video files, converts them into image sequences, and then processes these sequences to recognize specific human activities or events. This is primarily for computer vision researchers and AI model trainers focused on video analysis.
122 stars. No commits in the last 6 months.
Use this if you are a machine learning researcher or engineer looking to train and evaluate 3D ResNet models for action recognition on large video datasets like ActivityNet or Kinetics.
Not ideal if you want a tool to classify your own videos using pre-trained models without needing to train new ones or if you are not comfortable with command-line operations and PyTorch/Torch.
Stars
122
Forks
21
Language
Lua
License
MIT
Category
Last pushed
Nov 29, 2017
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/kenshohara/3D-ResNets"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
talmolab/sleap
A deep learning framework for multi-animal pose tracking.
kennymckormick/pyskl
A toolbox for skeleton-based action recognition.
open-mmlab/mmaction2
OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark
jgraving/DeepPoseKit
a toolkit for pose estimation using deep learning
kenshohara/3D-ResNets-PyTorch
3D ResNets for Action Recognition (CVPR 2018)