Português English
Contato

Proposta de Tese Aasim Khurshid


Detalhes do Evento


Aluno: Aasim Khurshid
Orientador: Prof. Dr. Jacob Scharcanski

Título: Adaptive Face Tracking Based on Online Learning
Linha de Pesquisa: Processamento de Imagens e Visão Computacional e Reconhecimento de Padrões

Data: 12/07/2018
Horário: 14h
Local: Prédio 43412 –  Sala 215 (sala de videoconferência) do Instituto de Informática

Banca Examinadora:
-Prof. Dr. Edison Pignaton De Freitas (UFRGS)
-Prof. Dr. Maciel Zortea (IBM Reseach, Brazil – por videoconferência)
-Prof. Dr. John Soldera (IFFAR)

Presidente da Banca: Prof. Dr. Jacob Scharcanski

Abstract: Object tracking can be used to localize objects in scenes, and also can be used for locating changes in the objects appearance or shape over time. Most of the available object tracking methods tend to perform satisfactorily in controlled environments, but tend to fail when the objects appearance or shape changes, or even when the illumination changes (e.g. when tracking non-rigid objects such as human face). Also, in many available tracking methods, the tracking error tends to increase indefinitely when the target is missed. Therefore, tracking the target objects in long (uninterrupted) video sequences tends to be quite challenging for these methods. This thesis proposal introduces a two variations of an object tracking algorithm which are applied to face tracking. The proposed methods are based on feature learning techniques that utilizes the useful data accumulated during the object tracking and implement incremental learning framework. To accumulate the training data, the quality of the test sample is checked before its utilization in the incremental and online training scheme. Also, a novel error prediction scheme is proposed that is capable to estimate the tracking error during the execution of the tracking algorithm. Furthermore, an improvement in the Constrained Local Model (CLM) is proposed that utilize the training data to assign weights to the landmarks based on their consistency. These weights are used in the CLM search process to improve CLM search optimization process. The experimental results show that the proposed tracking method (both variants) perform better than the comparative state of the art methods in terms of Root Mean Squared Error (RMSE) and Center Location Error (CLE). In order to prove the efficiency of the proposed techniques, an application in yawning detection is presented.

Keywords: Visual object tracking. Face tracking. Facial feature tracking. Tracking error predictor. Online learning. Incremental PCA/ICA. Dictionary learning.