Scenes Detection into Semantically Complex Segments

• Grant number: 11/05238-0
• Support Opportunities: Scholarships in Brazil - Master
• Start date: March 01, 2012
• End date: March 31, 2013
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Rudinei Goularte
• Grantee: Bruno Lorenço Lopes
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

This project aims to develop a video's segmentation technique that is able to identify scenes into semantically complex segments. In this kind of segments scene's transitions occurs when the segment's subject or topic changes. The proposal is to investigate the Bag of Features technique in order to semantically categorize the segments of a video. The semantics achieved by this technique is latent and only relates to the visual characteristics of the video, clustering together visually similar segments. Thus, it is proposed to investigate ways to extend the Bag of Features technique with multimodality, processing also the features that come from other media present in the video (as audio and closed captioning). In this way, it will be possible to recognize visually different neighbors segments, which, however, have the same subject, as belonging to the same scene.

Academic Publications

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

LOPES, Bruno Lorenço. Detection of scenes in semantically complex segments. 2014. Master's Dissertation - Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB) São Carlos.