Stereo Matching as a Nearest-Neighbor Problem
IEEE Transactions on Pattern Analysis and Machine Intelligence
Advances in Computational Stereo
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
Search is not inherent in the correspondence problem. We propose a representation of images, called intrinsic curves, that combines the ideas of associative storage of images with connectedness of the representation: intrinsic curves are the paths that a set of local image descriptors trace as an image scanline is traversed from left to right. Curves become surfaces when full images are considered instead of scanlines. Because only the path in the space of descriptors is used for matching, intrinsic curves lose track of space, and are invariant with respect to disparity under ideal circumstances. Establishing stereo correspondences then becomes a trivial lookup problem. We also show how to use intrinsic curves to match real images in the presence of noise, brightness bias, contrast fluctuations, and moderate geometric distortion, and we show how intrinsic curves can be used to deal with image ambiguity and occlusions. We carry out experiments on single-scanline matching to prove the feasibility of the approach and illustrate its main features.