Abstract
Action recognition has been a growing research topic in computer vision due to its great potentials for real-world applications. In this paper, we develop an effective action recognition approach based on salient object detection and propose a new feature descriptor to represent the changes of edge orientation. Firstly, we detect salient objects from each frame of a video sequence and generate edge maps for those detected salient objects. Then, we extract features on developed edge maps, using a combination of proposed Histogram of Changing Edge Orientation (HCEO) feature descriptor and existing Histogram of Optical Flow (HOF) feature descriptor. Finally, supervised multi-class support vector machine (SVM) classifier is used for recognizing various actions. The experiments were carried out on the standard UCF-Sports action dataset. As experimental results, our proposed action recognition approach is achieved with a significant improvement in recognition accuracy.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
