A Survey of Image Based Illumination Model
Chen Shen · Chinese Journal of Computers · 2000
This paper surveys the recent techniques for image based re lighting, i.e., rendering images with novel lighting conditions from a collection of images. Unlike image based rendering where static lighting condition is generally applied, image based re lighting allows significant lighting effects. Based on the plenoptic function, which is the theoretical basis for many image based rendering approaches, this paper argues that image based re lighting can be regarded as the process to sample, reconstruct, synthesize and re sample the plenoptic function. Authors classify previous techniques on image based re lighting into three categories according to what kind of illumination model is used: rendering with the traditional illumination model, rendering with the image based illumination model, and rendering without illumination model. Authors further suggest in the paper that the future work on image based re lighting should focus on image based illumination model. Finally, this paper briefly introduces a new image based illumination model to add new objects into an IBR system.