Recursive Edge-Aware Filters for Stereo Matching

Cevahir Cigla; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2015, pp. 27-34

Abstract


In this study, taxonomy of recursive edge-aware filters (REAF) is provided, with the introduction of new approaches to the state-of-the-art. The one tap recursive filters are classified according to recursion rate calculation, recursion type involving normalized-un-normalized recursion and the unification of reverse directions that is also valid for higher order recursions. In that manner, eight types of edge-aware recursive filters are defined, where only three of them are addressed in literature so far. Comprehensive analyses are provided based on computational complexity and filter characteristics which affect the use of such filters for various applications. In order to compare the capabilities of these filters, stereo matching, as the most common application area of edge-aware filters, is considered and extensive experiments are provided through well known datasets. The evaluation is conducted on large number of stereo pairs with independent parameter optimization of each filter providing fair comparison. According to the experimental results, advantages of un-normalized recursion for matching accuracy and sequential integration of reverse directions for execution speed are illustrated as important conclusions for future directions of REAFs.

Related Material


[pdf]
[bibtex]
@InProceedings{Cigla_2015_CVPR_Workshops,
author = {Cigla, Cevahir},
title = {Recursive Edge-Aware Filters for Stereo Matching},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2015}
}