Structural pattern recognition and intermediate representations

• Grant number: 19/06254-0
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: May 01, 2019
• End date: January 31, 2020
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Roberto Marcondes Cesar Junior
• Grantee: Vinicius Ferraris Pignataro Mazzei Albert
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 15/22308-2 - Intermediate representations in Computational Science for knowledge discovery, AP.TEM

Abstract

In recent years, research on 3D shape recognition(medical imaging) and actions (video) has been developing rapidly,especially for dynamic environments or video sequences. The detectionand recognition of salient elements may play aimportant in such systems. The concept of "salient" can beapplied to different actors or objects involved in dynamic scenes,in terms of both structural and temporal information. Thesalient elements represent one of the main focuses of thisproject. The basic idea is to extract the visual elements based onin a hierarchical structure of three layers: 1 - extraction ofsalient elements, 2 - learning of important characteristics and 3- detection of salient elements for the recognition of actions,respectively. The integration of structural relationships to detect,trace and recognize the salient elements will be exploited to therecognition of actions, since thistype of track can be considered more stable than theappearance of the object, especially for the overall improvement of the methodperformance. The pattern recognition approaches mentioned in theThematic project will be explored. Thus, the goal of the project is to develop a method of structural recognition on 3D medical image data or video sequences using intermediate representations. We intend to explore salient elements as the basis of such intermediate relations.

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)

FERRARIS, VINICIUS; DOBIGEON, NICOLAS; CAVALCANTI, YANNA; OBERLIN, THOMAS; CHABERT, MARIE; IEEE. UNSUPERVISED CHANGE DETECTION FOR MULTIMODAL REMOTE SENSING IMAGES VIA COUPLED DICTIONARY LEARNING AND SPARSE CODING. 2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, v. N/A, p. 5-pg., 2020-01-01. (19/06254-0)