human-action-recognition

custom human activity recognition modules by pose estimation and cascaded inference using sklearn API

Keras implementation of Human Action Recognition for the data set State Farm Distracted Driver Detection (Kaggle)

Source code for "Learning Graph Convolutional Network for Skeleton-based Human Action Recognition by Neural Searching", AAAI2020

This repository allows you to classify 40 different human actions. Pose detection, estimation and classification is also performed. Poses are classified into sitting, upright and lying down.

deep learning sex position classifier

A Comprehensive Tutorial on Video Modeling

[TPAMI 2020] "Privacy-Preserving Deep Action Recognition: An Adversarial Learning Framework and A New Dataset" by Zhenyu Wu, Haotao Wang, Zhaowen Wang, Hailin Jin, and Zhangyang Wang

Computer Vision Project : Action Recognition on UCF101 Dataset

Activity Recognition using Temporal Optical Flow Convolutional Features and Multi-Layer LSTM

Surveillance Perspective Human Action Recognition Dataset: 7759 Videos from 14 Action Classes, aggregated from multiple sources, all cropped spatio-temporally and filmed from a surveillance-camera like position.

MSR Action Recognition Datasets and Codes

Human Activity Recognition Research Repository

Deep learning model that predicts human action in a given video feed using pose estimation

This repository contains the MPOSE2021 Dataset for short-time pose-based Human Action Recognition (HAR).

This repository provides implementation of a baseline method and our proposed methods for efficient Skeleton-based Human Action Recognition.

A human action dataset collected from Elder Scrolls V: Skyrim

Implementation of some popular skeleton based Human Action Recognition methods basis on Deep Neural Networks.

Code for HAR-GCNN: Deep Graph CNNs for Human Activity Recognition From Highly Unlabeled Mobile Sensor Data, IEEE PerCom CoMoRea 2022

human action recognition -- pipeline of improved dense trajectory

A skeleton-based real-time online action recognition project, classifying and recognizing base on framewise joints, which can be used for safety surveilence.

his is a human action recognition(HAR) project based on CNNs and Tensorflow using a pretrained model.

This python opencv code is used to segment the human object from the video frame

Source code of experiments performed in paper: Human Action Recognition in Videos Based on Spatiotemporal Features and Bag-of-Poses

Synthetically Generated Surveillance Perspective Human Action Recognition Dataset: 6901 Videos from 10 action classes, made by a 3D Simulation, all cropped spatio-temporally and filmed from a surveillance-camera like position.

Human Activity Recognition UCI Dataset, final score 0.97196

This includes a novel method to measure the quality of the actions performed in Olympic weightlifting using human action recognition in videos. Human action recognition is a well-studied problem in computer vision and on the other hand action quality assessment is researched and experimented comparatively low. This is due to the lack of datasets that can be used to assess the quality of actions. In this research, we introduce a method to assess player techniques in weightlifting by using skeleton-based human action recognition. Furthermore, we introduce a new video dataset for action recognition in weightlifting which is annotated to frame level. We intended to develop a viable automated scoring system through action recognition that would be beneficial in the sports industry.

Project to explore a deep learning solution to a computer vision problem. Human action recognition has become increasingly popular. This project implements a deep RNN to detect seizures.

A system for Human Action Recognition that uses the scale and body orientation invariant Skeletal Quads representation, with an LSTM network

Implementation of CNN-Based Model for Online Action Recognition

This is an effort to provide different approaches towards human action recognition from video. A method to perform data augmentation on skeletal data so as to achieve a view independent recognition approach is included.