Anomaly detection using divergence measures and feature decomposition in video

• Grant number: 15/13504-2
• Support Opportunities: Scholarships abroad - Research
• Start date: February 01, 2016
• End date: January 31, 2017
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Moacir Antonelli Ponti
• Grantee: Moacir Antonelli Ponti
• Host Investigator: Teófilo Emidio de Campos
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: University of Surrey, England

Abstract

The analysis of video sequences through features often considers each frame as an static image, for which interest points are detected and some feature extraction is carried out. Afterwards, a temporal model is applied in order to learn from the features. However there are few studies using both spatial and temporal information available, i.e. the spatial information within a frame and the temporal information on how the features change in time. Also, it is possible to explore both time and spatial features, detecting domain anomalies via classifier divergence analysis. This project aims to analyze features from divergence analysis between classifier and using multidimensional decomposition of the video sequence, which will possibility better understand changes in the behavior through time. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

PONTI, MOACIR; KITTLER, JOSEF; RIVA, MATEUS; DE CAMPOS, TEOFILO; ZOR, CEMRE. A decision cognizant Kullback-Leibler divergence. PATTERN RECOGNITION, v. 61, n. SI, p. 470-478, JAN 2017. (15/13504-2, 15/24652-2)

PONTI, MOACIR; BET, PATRICIA; OLIVEIRA, CAROLINE L.; CASTRO, PAULA C.. Better than counting seconds: Identifying fallers among healthy elderly using fusion of accelerometer features and dual-task Timed Up and Go. PLoS One, v. 12, n. 4, APR 27 2017. (13/07375-0, 15/09715-8, 15/13504-2, 16/10982-3, 14/21858-6)

PONTI, MOACIR; KITTLER, JOSEF; RIVA, MATEUS; DE CAMPOS, TEOFILO; ZOR, CEMRE. A decision cognizant Kullback-Leibler divergence. PATTERN RECOGNITION, v. 61, p. 9-pg., 2017-01-01. (15/13504-2, 15/24652-2)