14 February 2015 SubPatch: random kd-tree on a sub-sampled patch set for nearest neighbor field estimation
Author Affiliations +
Proceedings Volume 9445, Seventh International Conference on Machine Vision (ICMV 2014); 944514 (2015) https://doi.org/10.1117/12.2180577
Event: Seventh International Conference on Machine Vision (ICMV 2014), 2014, Milan, Italy
Abstract
We propose a new method to compute the approximate nearest-neighbors field (ANNF) between image pairs using random kd-tree and patch set sub-sampling. By exploiting image coherence we demonstrate that it is possible to reduce the number of patches on which we compute the ANNF, while maintaining high overall accuracy on the final result. Information on missing patches is then recovered by interpolation and propagation of good matches. The introduction of the sub-sampling factor on patch sets also allows for setting the desired trade off between accuracy and speed, providing a flexibility that lacks in state-of-the-art methods. Tests conducted on a public database prove that our algorithm achieves superior performance with respect to PatchMatch (PM) and Coherence Sensitivity Hashing (CSH) algorithms in a comparable computational time.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Fabrizio Pedersoli, Fabrizio Pedersoli, Sergio Benini, Sergio Benini, Nicola Adami, Nicola Adami, Masahiro Okuda, Masahiro Okuda, Riccardo Leonardi, Riccardo Leonardi, } "SubPatch: random kd-tree on a sub-sampled patch set for nearest neighbor field estimation", Proc. SPIE 9445, Seventh International Conference on Machine Vision (ICMV 2014), 944514 (14 February 2015); doi: 10.1117/12.2180577; https://doi.org/10.1117/12.2180577
PROCEEDINGS
5 PAGES


SHARE
Back to Top