The research titled "Video Content Analysis and Classification Based on Human Activity Recognition" is published by Springer Nature. It is also a part of a book series Communications in Computer and Information Science ((CCIS,volume 2376)).

The authors are Hari K.C, Prof Dr. Manish Pokharel and Dr. Sushil Shrestha.

Abstract: Video content analysis holds significant potential in computer vision tasks for the automated understanding and interpretation of human motion and activities. This paper aims to introduce an intelligent framework to recognize human motion and activities within video data. By leveraging advanced techniques in feature extraction, temporal modeling and deep learning, a comprehensive approach that aims to accurately categorize and interpret various human activities is presented in this study. The combination of Convolution layers and Bidirectional Long Short-Term Memory Layers is used to extract the spatial and temporal features of the video enabling the detection of motion patterns and activity sequences. Video classification and Activity Recognition (VCHAR) is developed and trained using the University of Central Florida 101 Classes (UCF101) dataset. This dataset consists of 101 classes of activities such as archery, surfing, typing, swing and so on. The model is tested on day-to-day human activities videos. The study provides the classification of videos based on the activities within the video. The result of this study is compared with previous benchmark results and it showed better accuracy of 87.67%. The finding of this study contributes to improving surveillance security, sports activity, intrusion detection and robotics. Sport media companies, automation Industry, education sector and researchers will be benefitted from this type of study.

The link detail: https://link.springer.com/chapter/10.1007/978-3-031-81935-3_23