Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 44477 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Kasaei, Shohreh
- Abstract:
- In recent years, ability of computers to recognize human actions, because of numerousapplications, has attracted scientists. Surveillancesystems in house, work and public places, human computer interaction, study of human movement problems, remote supervision of ill or old people and sport training are only some of the applications. In this thesis 10 actions are considered. These actions are Walking, Running, Galloping side, Bending, Jump jacking, Jumping, Jumping in place, Skipping, Waving one hand and Waving two hands. All actions exist in Weisemann dataset so this dataset is used as training and testing dataset. Here important objectives are recognising human action so that it is indipendent of subject, independent of speed of actions, robustness against noise, good capacity of number of actions and no requirement for large amount of traing data. In this work the procedure is as fallows: First the silhouettes are extracted from the input videos. Next the locational-temporal featues are extracted so that it is robust against rotation, zoom and trasfer of subjects. After that an undirected graphical model that works discriminatively is traind using trainig data. Finally the trained model is tested using testing data. Evaluation of the method is done by comparing confusion matrices of the model and latest published papers. The method has 100% accuracy with 6 actions and 89% accuracy with all 10 actions. So the method proposed here has good accuracy comparing to other graphical methods used in this field. Complexity of the method inlearning phase is of orderO(n^3) and in testing phase is of order O(n)
- Keywords:
- Human Action Categorization ; Silhouette ; Graphic Model ; Surveillance System ; Locational-Temporal Feature
Digital Object List
-
محتواي کتاب
- view
Bookmark