Multiview object detection using keygraphs

• Grant number: 12/09741-0
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: September 01, 2012
• End date: August 31, 2015
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Roberto Marcondes Cesar Junior
• Grantee: Henrique Morimitsu
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated scholarship(s): 13/08258-7 - High-level event characterization using graphs, BE.EP.DR

Abstract

This document presents a project for a doctorate in computer vision for thestudent Henrique Morimitsu, advised by Professor Roberto Marcondes Cesar Juniorfrom the Computer Science Department of the Institute of Mathematics and Statisticsof the University of São Paulo. This project will count with the collaboration of ProfessorIsabelle Bloch (Télécom ParisTech - Paris).This project consists of the study and development of structural methods formultiview object detection. In order to accomplish such task, we intend touse the keygraphs object detection method.Keygraphs are a generalization of the traditional keypoint object detectionmethod. This method was studied by the student during his master's courseand presented good results for singleview object detection. In this doctorate,we intend to study ways of extending it in order to to be able to performmultiview object detection. Besides the extension of the method, we also intendto study how to integrate methods of machine learning to obtain betterresults.

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)

MORIMITSU, HENRIQUE; BLOCH, ISABELLE; CESAR-, ROBERTO M., JR.. Exploring structure for long-term tracking of multiple objects in sports videos. COMPUTER VISION AND IMAGE UNDERSTANDING, v. 159, p. 16-pg., 2017-06-01. (14/50135-2, 11/50761-2, 13/08258-7, 12/09741-0)

MORIMITSU, HENRIQUE; BLOCH, ISABELLE; CESAR-, JR., ROBERTO M.. Exploring structure for long-term tracking of multiple objects in sports videos. COMPUTER VISION AND IMAGE UNDERSTANDING, v. 159, n. SI, p. 89-104, JUN 2017. (14/50135-2, 13/08258-7, 12/09741-0, 11/50761-2)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

MORIMITSU, Henrique. A graph-based approach for online multi-object tracking in structured videos with an application to action recognition. 2015. Doctoral Thesis - Universidade de São Paulo (USP). Instituto de Matemática e Estatística (IME/SBI) São Paulo.