Person Recognition from Activity using Bag of Words
Vidhya.V.S.Nair1, Subha V2
1Vidhya.V.S.Nair, Department of Electronics and Communication, SCT College of Engineering, Pappanamcode (Kerala), India.
2Subha V, Department of Electronics and Communication, SCT College of Engineering, Pappanamcode (Kerala), India.
Manuscript received on 13 June 2016 | Revised Manuscript received on 20 June 2016 | Manuscript Published on 30 June 2016 | PP: 78-81 | Volume-5 Issue-5, June 2016 | Retrieval Number: E4610065516/16©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In this paper the discriminant pattern hidden in the way of doing an activity for every person is explored. This pattern can be utilized for person recognition purpose in uncontrolled scenarios unlike finger print, iris, retina etc. (based on physical biometrics). This method is based on single video camera based data. From the video of various activities, background subtraction is done to remove insignificant data. From the binary video obtained after background subtraction structural tensor based features are detected and extracted. The extracted features defines the variation from the mean position are then clustered by means of k-means clustering. Histogram of cluster centroids is calculated using Bag Of Words (BOW) and classified by category classifier. Histogram of input video action sequence is compared with each of dataset and predicts the category, which corresponds to the label of person.
Keywords: Activity Based Identification, Background Subtraction, Silhouette, Structural Tensor, Bag Of Words, Category Classifier, Structured Support Vector Machine.
Scope of the Article: Pattern Recognition