An attentional model for videos classification

• Grant number: 18/10027-7
• Support Opportunities: Scholarships in Brazil - Master
• Start date: January 01, 2019
• End date: September 30, 2020
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Gerberth Adín Ramírez Rivera
• Grantee: Darley Freire Barreto
• Host Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

An important property of human perception is that one does not tend to process a whole scene in its entirety at once, for example, if we want to tell which model is a car, we do not need to look at people or animals in the environment. Likewise, when doing classification in pictures, we need to “pay attention” in certain parts of the image. Attention is an important component of neural networks that adapts them to understand sequences such, as video, voice, text, etc. In recent years, researchers have using attention to yield state-of-art models for tasks such, as image classification, image captioning, image generation, object recognition, action recognition, image segmentation, video captioning, machine translation, visual question answering, etc. One big challenge is that the majority of attention-based models focus on Natural Language or related tasks, and video classification is not that much explored. Thus in this Masters project, we propose to evaluate state-of-art models and to create a novel model to classify video based on attention. (AU)

Academic Publications

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

BARRETO, Darley Freire. Um modelo de atenção visual hierárquico. 2021. Master's Dissertation - Universidade Estadual de Campinas (UNICAMP). Instituto de Computação Campinas, SP.