Sparse Recursive Cost Aggregation Towards O(1) Complexity Local Stereo Matching

2015-05-19
The complexity of the local stereo matching methods mainly increases with disparity search range and cost aggregation step. Joint elimination of the those complexity factors is a challenging task as a consequence of the contradicting nature of the methods attacking the reduction on the complexity factors. In this paper, that challenge is addressed and for the disparity search range reducing approaches, an efficient cost aggregation method is proposed by reformulating the filtering scheme of the recursive edge-aware filters which have been proved to be efficient approaches for cost aggregation. The proposed method is exploited by a hierarchical stereo matching approach. In that manner, fixed number of disparity candidates are tested for each pixel, regardless of the search space and the cost aggregation for each candidate is performed with constant complexity. The experimental results validate that the proposed approach has linear complexity with the image size and show that in practice it speeds up the recursive approaches almost four times with 0.01-0.96% decrease in matching accuracy. Compared to the state-of-the-art techniques, the proposed method is possibly the fastest approach with a competitive accuracy based on Middlebury benchmarking.

Suggestions

Sparse recursive filtering for O 1 stereo matching
Gürbüz, Yeti Ziya; Alatan, Abdullah Aydın (2015-09-30)
Recursive edge-aware filters have been proved to be one of the most efficient approaches for cost aggregation in stereo matching. However, disparity search space dependency, as a result of full search, is the bottle-neck of these local techniques that prevent further reduction in computation. In this paper, the cost aggregation and correspondence search problems are re-formulated to enable adaptive search for each pixel during recursive operations that provides significant reduction in computational complex...
Occlusion aware stereo matching with o(1) complexity /
Gürbüz, Yeti Ziya; Alatan, Abdullah Aydın; Department of Electrical and Electronics Engineering (2014)
The problem of joint reduction of computational complexities of local stereo matching methods due to both cost aggregation step and correspondence search range is addressed and a novel hierarchical stereo matching algorithm is presented. The proposed approach exploits edge aware recursive volume filtering with a reduction on correspondence search range. The fastest state-of-the-art edge aware recursive filters are modified so that they become applicable to the methods to reduce the complexity in corresponde...
Edge-Aware Stereo Matching with O(1) Complexity
Cigla, Cevahir; Alatan, Abdullah Aydın (2012-01-26)
In this paper, a novel local stereo matching algorithm is introduced, providing precise disparity maps with low computational complexity. Following the common steps of local matching methods, namely cost calculation, aggregation, minimization and occlusion handling; the time consuming intensity dependent aggregation procedure is improved in terms of both speed and precision. For this purpose, a novel approach, denoted as permeability filtering (PF), is introduced, engaging computationally efficient two pass...
Segment-based stereo-matching via plane and angle sweeping
Cigla, Cevahir; Zabulis, Xenophon; Alatan, Abdullah Aydın (2007-01-01)
A novel approach for segment-based stereo matching problem is presented, based on a modified plane-sweeping strategy. The space is initially divided into planes that are located at different depth levels via plane sweeping by the help of region-wise planarity assumption for the scene. Over-segmented homogenous color regions are utilized for defining planar segment boundaries and plane equations are determined by angle sweeping at different planes. The robustness of depth map estimates is improved by warping...
Cluster searching strategies for collaborative recommendation systems
Altıngövde, İsmail Sengör; Ulusoy, Ozgur (2013-05-01)
In-memory nearest neighbor computation is a typical collaborative filtering approach for high recommendation accuracy. However, this approach is not scalable given the huge number of customers and items in typical commercial applications. Cluster-based collaborative filtering techniques can be a remedy for the efficiency problem, but they usually provide relatively lower accuracy figures, since they may become over-generalized and produce less-personalized recommendations. Our research explores an individua...
Citation Formats
Y. Z. Gürbüz and A. A. Alatan, “Sparse Recursive Cost Aggregation Towards O(1) Complexity Local Stereo Matching,” 2015, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/53548.