Activity recognition based on event probability sequence and key point detecting

Xiaoxing Li; Yi Yang

doi:10.1117/12.913484

30 September 2011 Activity recognition based on event probability sequence and key point detecting

Xiaoxing Li, Yi Yang

Proceedings Volume 8285, International Conference on Graphic and Image Processing (ICGIP 2011); 828524 (2011) https://doi.org/10.1117/12.913484
Event: 2011 International Conference on Graphic and Image Processing, 2011, Cairo, Egypt

Abstract

In this paper we propose a novel approach for video activity recognition based on event probability sequence (EPS) and key point description. Meaningful key points are first detected and trajectories are compressed according to these key points. Then EPSs are calculated for each refined trajectory to represent characteristics of trajectories. Dynamic time warping (DTW) algorithm is used to metric similarity between testing trajectories and training trajectories, and match them efficiently. Also particle swarm optimization (PSO) is included to improve the learning process of hidden Markov model (HMM). Experiments show that this method achieves higher recognition rate using less time on the UCF human action data set.

Citation Download Citation

Xiaoxing Li and Yi Yang "Activity recognition based on event probability sequence and key point detecting", Proc. SPIE 8285, International Conference on Graphic and Image Processing (ICGIP 2011), 828524 (30 September 2011); https://doi.org/10.1117/12.913484

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available