I begin PhD studies and my research based on integral imaging. As a part of my research, I want to perform simulations of the Integral Imaging (II) process. For this I need mathematical description of the forward process (ray (or wavefront) propagation from the object to the array of the elemental images). Can some body recommend relevant references? Thank you in advance.
Vladimir,
I'm not completely sure that I understand what you intend to do. Are you looking for a way to produce integral images from a given camera system and a defined scene, or render virtual views from a already availble integral imaging content? Either way, you might find some useful information about the first option (and some of the second) in my thesis "Synthesis, Coding, and Evaluation of 3D Images Based on Integral Imaging", which you can download from http://www.miun.se/en/personnel/RogerOlsson/.
Good luck with your integral imaging research and don't hesitate to contact me if you have any questions related to my work.
BR,
Roger
Thank you for your response, Sathish and Roger.
In the first step of the research, I want to simulate step of creating array of the elemental images from some virtual 3D scene, for example two virtual dices. So I also stuck in the step of generating of the correct 3D scene in MATLAB. I will be glad to get advice (and relevant links) on this matter.
Thank you again.
Vladimir,
There are a few ways you can go about rendering the set of elemental images. Object based rendering (rasterization) or raytracing is of course the main design choice to make. Consider the lightfield/plenoptic camera elements to be part of the world definition or not is another aspect, especially in the raytracing case. Constructing the scene with an easy to use 3D-modelling software or an explicit scene description language is a question about accuracy and repeatability. Feel free to read about how I navigated through these issues in my thesis or in my paper "An Interactive Ray-tracing based simulation environment for generating integral imaging video sequences" (2005), both of which you can download from the above link. A slightly different approach to raytracing integral imaging content have also been pursued in "Photorealistic integral photography using a ray-traced model of capturing optics" (2006) by Athineos et al.. For work in the rasterization branch of rendering the paper "Multiple viewpoint rendering" (1998) by Michael Halle is the paper to start with. Using Google scholar you can then find a wealth of work within that area: http://scholar.google.se/scholar?cites=15919567511517336804&as_sdt=2005&sciodt=0,5&hl=sv.
Using my approach your Dice example would be straightforward divided into defining the scene, defining a camera model that describes your image capture system, and render. Decoupling camera from scene speeds up rendering and allows for exchanging camera models and scene definitions independently.
BR,
Roger
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Computer Graphics and Animation
- 3D