Structure Models for Image-Assisted Geometry Measurement in Plenoptic Sampling

We present a signal-processing framework for image-assisted geometry measurement in the image-based rendering (IBR). We study the utilized geometry information and estimating minimum sampling rate of the IBR. Our method combines decomposing a complex scene geometry into a collection of simpler structures on a block-by-block basis. The automatic simpler structure selection can be interactively refined by detected single salient points. In this manner, we reduce the spectral analysis problem of an irregular object to that of a simpler structure. Predictions on the frequency content can then be used to control the sampling rate. This extends previous work in which the IBR sampling is analyzed and estimated for nonuniform sampling. Extensive experimental evaluation demonstrates that our geometry simplification method significantly outperforms competing algorithms. Additionally, the minimum sampling rate of the IBR necessary for alias-free rendering will be reduced as the number of simpler structures increases.