A survey on vision based action recognition in machine learning
LE3 .A278 2022
2022
Zhang, Haiyi
Acadia University
Bachelor of Science
Honours
Computer Science
Action recognition is a field derived from the rapid development of computer vision, which is a field within machine learning. Computer vision refers to the use of computers to extract data from images, videos, or other visual inputs, and the ability to analyze and organize them. The ability to recognize objects within a picture is a typical example of computer vision. For example, Flower Recognition is software developed by Microsoft to identify different types of flowers within a photograph, and action recognition does the same thing as the flower recognition software, extracting the action of a person in a video and identifying the type of that action. Action recognition consists of two main steps, extracting action features and classifying the action. This paper will focus on the two steps that make up action recognition, as well as common action recognition applications, datasets, and the challenges faced.
The author retains copyright in this thesis. Any substantial copying or any other actions that exceed fair dealing or other exceptions in the Copyright Act require the permission of the author.
https://scholar.acadiau.ca/islandora/object/theses:3783