A template tracker is a tracker based on matching a pre-initialised view of an object with the object's view in an image sequence. Using an error function, the intensity difference between the template view and the templated region in the image is measured. This error measure is used as the basis for a template alignment algorithm that will adjust the template's pose to more accurately register the template view with the view of the object in the image. Some significant problems present themselves with this simple tracker. Extraneous, or non-object, pixels within the template boundaries can cause bias in the registration of the template. Partial occlusions of the object's view in the image can also cause serious bias in the template's pose. Beyond simple occlusions there are transits of occlusions across an object. Occlusion transits are significant because over time they can occlude the entire object in an incremental fashion. If initially the template view is not completely known this kind of occlusion can easily cause a total tracking failure for an object. In this thesis three enhancements of the basic template tracker are proposed: Ownership Masks, Cooperative Templates, and Evolving Views. Ownership Masks are aimed at eliminating the extraneous pixels from the template view. Cooperative templates are used to separate the intensity probabilities when more than one template covers a pixel. Building upon both Ownership Masks and Cooperative Templates, Evolving Views update the template views when occlusion transits are a problem. With these enhancements we have been able to increase the accuracy of tracking objects where large portions of a template contain background pixels. Also occlusions and some types of unocclusions can be detected and discriminated. Finally, some failures in the basic tracker due to occlusion transits have been overcome.

A template tracker is a tracker based on matching a pre-initialised view of an object with the object's view in an image sequence. Using an error function, the intensity difference between the template view and the templated region in the image is measured. This error measure is used as the basis for a template alignment algorithm that will adjust the template's pose to more accurately register the template view with the view of the object in the image. Some significant problems present themselves with this simple tracker. Extraneous, or non-object, pixels within the template boundaries can cause bias in the registration of the template. Partial occlusions of the object's view in the image can also cause serious bias in the template's pose. Beyond simple occlusions there are transits of occlusions across an object. Occlusion transits are significant because over time they can occlude the entire object in an incremental fashion. If initially the template view is not completely known this kind of occlusion can easily cause a total tracking failure for an object. In this thesis three enhancements of the basic template tracker are proposed: Ownership Masks, Cooperative Templates, and Evolving Views. Ownership Masks are aimed at eliminating the extraneous pixels from the template view. Cooperative templates are used to separate the intensity probabilities when more than one template covers a pixel. Building upon both Ownership Masks and Cooperative Templates, Evolving Views update the template views when occlusion transits are a problem. With these enhancements we have been able to increase the accuracy of tracking objects where large portions of a template contain background pixels. Also occlusions and some types of unocclusions can be detected and discriminated. Finally, some failures in the basic tracker due to occlusion transits have been overcome.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-06-07T14:26:02Z No. of bitstreams: 1 helenadealmeidamaia.pdf: 3132814 bytes, checksum: d46a470b453ec6ba11362abaeac3a42c (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-06-07T14:56:50Z (GMT) No. of bitstreams: 1 helenadealmeidamaia.pdf: 3132814 bytes, checksum: d46a470b453ec6ba11362abaeac3a42c (MD5) / Made available in DSpace on 2017-06-07T14:56:50Z (GMT). No. of bitstreams: 1 helenadealmeidamaia.pdf: 3132814 bytes, checksum: d46a470b453ec6ba11362abaeac3a42c (MD5) Previous issue date: 2016-03-18 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Nos últimos anos, o rastreador TLD (Tracking-Learning-Detection) se destacou por combinar um método de rastreamento através do movimento aparente e um método de detecção para o problema de rastreamento de objetos em vídeos. O detector identifica o objeto pelas aparências supostamente confirmadas. O rastreador insere novas aparências no modelo do detector estimando o movimento aparente. A integração das duas respostas é realizada através da mesma métrica de similaridade utilizada pelo detector que pode levar a uma decisão enviesada. Neste trabalho, é proposto um framework para métodos baseados em múltiplos rastreadores onde o componente responsável pela integração das respostas é independente dos rastreadores. Este componente é denominado mediador. Seguindo este framework, um novo método é proposto para integrar o rastreador por movimento e o detector do rastreador TLD pela combinação das suas estimativas. Os resultados mostram que, quando a integração é independente das métricas de ambos os rastreadores, a performance é melhorada para objetos com significativas variações de aparência durante o vídeo. / On the problem of tracking objects in videos, a recent and distinguished approach combining tracking and detection methods is the TLD (Tracking-Learning-Detection) framework. The detector identifies the object by its supposedly confirmed appearances. The tracker inserts new appearances into the model using apparent motion. Their outcomes are integrated by using the same similarity metric of the detector which, in our point of view, leads to biased results. In our work, we propose a framework for generic multitracker methods where the component responsible for the integration is independent from the trackers. We call this component as mediator. Using this framework, we propose a new method for integrating the motion tracker and detector from TLD by combining their estimations. Our results show that when the integration is independent of both tracker/detector metrics, the overall tracking is improved for objects with high appearance variations throughout the video.