Publicação

Light field processor: a lytro illum imaging application

Ver documento

Detalhes bibliográficos
Resumo:Light field imaging technology is at the intersection of three main research areas: computer graphics, computational photography and computer vision. This technology has the potential to allow functionalities that were previously impracticable, if not impossible, like refocusing photographic images after the capture or moving around in a VR scene produced by a real-time game engine, with 6DoF. Traditional photography produces one single output whenever a user presses the shot button. Light field photography may have several different outputs because it collects much more data about a scene. Thus it requires post-processing in order to extract any piece of useful information, like 2D images, and that is a characteristic feature that makes this technology substantially different from all others in the field of image making. Post processing means using a specialised application and, since this technology is still in its infancy, those applications are scarce. This context presented a good opportunity for such a development. Light Field Processor is the main outcome of this work. It is a computer application able to open and decode images from Lytro Illum light field cameras, which it may then store as a new file format (Decoded Light Field), proposed in this dissertation, for later use. It is able to extract 2D viewpoints, 2D maps of viewpoints or the microlens array, videos showing the intrinsic parallax of the light field and metadata, as well as do some basic image processing.
Autores principais:Silva, Agostinho
Assunto:Light field Image processing Computational photography Lytro illum Engenharia informática Desenvolvimento de software Aplicação multimédia Processamento de imagens Campo de luz Fotografia digital
Ano:2016
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:ISCTE
Idioma:inglês
Origem:Repositório ISCTE
Descrição
Resumo:Light field imaging technology is at the intersection of three main research areas: computer graphics, computational photography and computer vision. This technology has the potential to allow functionalities that were previously impracticable, if not impossible, like refocusing photographic images after the capture or moving around in a VR scene produced by a real-time game engine, with 6DoF. Traditional photography produces one single output whenever a user presses the shot button. Light field photography may have several different outputs because it collects much more data about a scene. Thus it requires post-processing in order to extract any piece of useful information, like 2D images, and that is a characteristic feature that makes this technology substantially different from all others in the field of image making. Post processing means using a specialised application and, since this technology is still in its infancy, those applications are scarce. This context presented a good opportunity for such a development. Light Field Processor is the main outcome of this work. It is a computer application able to open and decode images from Lytro Illum light field cameras, which it may then store as a new file format (Decoded Light Field), proposed in this dissertation, for later use. It is able to extract 2D viewpoints, 2D maps of viewpoints or the microlens array, videos showing the intrinsic parallax of the light field and metadata, as well as do some basic image processing.