UntrimmedNet

Action recognition system

A system for recognizing and detecting actions in untrimmed videos using a weakly supervised learning approach.

Weakly Supervised Action Recognition and Detection

GitHub

162 stars
18 watching
47 forks
Language: Matlab
last commit: about 6 years ago
Linked from 1 awesome list

action-detectionaction-recognitionweakly-supervised-learning

Backlinks from these awesome lists:

Related projects:

Repository Description Stars
zhiwu-huang/lienet This project uses deep learning and Lie group theory to recognize actions from skeleton data 64
wanglimin/tdd A tool for extracting features from videos using deep convolutional descriptors 104
lukereichold/visualactionkit An iOS framework for classifying human actions in videos using deep learning models 24
wangguanzhi/ladn An implementation of a deep learning model for facial makeup and removal transfer 181
jingzhang617/cod-rank-localize-and-segment Develops a system to detect, segment, and rank camouflaged objects in images. 74
naoto0804/cross-domain-detection Develops object detection algorithms to adapt to new domains with limited supervision 422
guillaume-chevalier/har-stacked-residual-bidir-lstms An implementation of a deep neural network architecture for Human Activity Recognition using stacked residual bidirectional LSTM cells with TensorFlow. 319
neuraxio/lstm-human-activity-recognition An LSTM-based human activity recognition system using smartphone sensor data. 3
shahroudy/nturgb-d A large-scale dataset for human action recognition 758
vowpalwabbit/vowpal_wabbit An online learning system that enables fast and scalable machine learning with techniques like active learning and contextual bandits. 8,485
xiaolonw/adversarial-frcnn A Caffe-based implementation of A-Fast-RCNN, a method for object detection using adversarial networks. 482
meituan-automl/lenna An AI-powered image detection system with language-based reasoning capabilities 78
liushenwenyuan/matlab_orc A Matlab implementation of a handwritten digit recognition system using neural networks. 50
juntang-zhuang/laddernet A deep learning implementation of a multi-path network architecture for medical image segmentation 139
ma-shamshiri/human-activity-recognition A project to classify human activities using data from accelerometer and gyroscope sensors 98