Estimation of disparity map of stereo image pairs using spatial domain local Gabor wavelet

Abstract

The stereo matching problem takes two images captured by nearby cameras and attempts to recover quantitative disparity information. Most of the existing stereo matching algorithms find it difficult to estimate disparity in the occlusion, discontinuities and textureless regions in the images. In the last few decades, a number of stereo matching methods have been proposed to overcome some of these problems. In the same line of thought, the authors propose a new feature-based stereo matching method, which consists of four basic steps -feature-based stereo correspondence, two-pass cost aggregation, disparity computation using winner-takes-all selection and finally, the disparity refinement. In the proposed method, local features of Gabor wavelet in spatial domain are used for matching cost computation and subsequently a cost aggregation step is implemented by combined use of the Kuwahara filter and the median filter. Experimental results on the Middlebury benchmark database shows that the proposed method outperforms many existing local stereo matching methods.

Asymmetric occlusion detection using linear regression and weight‐based filling for stereo disparity map estimation

“…Step 1 shows the methods by which the initial disparity map is estimated. For this, GW [12] and guided filter (GF) [13] are considered. Similarly for step 2, either the proposed occlusion detection and filling method (LASW) or LNDA [13] is used.…”

“…Fig. 7 shows the disparity maps estimated by our proposed method and the method proposed in [12]. These two methods differ only in the process of occlusion detection and filling, and all other steps of disparity map estimation remain same.…”

“…The output K ( p , q ) of Kuwahara filter for a pixel ( p , q ) is given by mean corresponding to a subregion having minimum standard deviation. This can be formulated as follows: where Details about cost computation and aggregation can be found in [12].…”

“…Comparison of disparity maps estimated by our proposed LASW method and the method proposed in[12]. Top to bottom -left images, ground truth disparity maps, disparity maps generated by our method, error image which indicate the bad pixels in the disparity maps, disparity maps generated by the method proposed in[12], and its corresponding bad pixel images. Left to right -Tsukuba, Venus, Teddy, and Cones images…”

Performance Analysis of Simulated Annealing Cooling Schedules in the Context of Dense Image Matching

“…Indeed, it is the fundamental precursor of 3D reconstruction from a stereo pair or image sequence. Dense image matching can be cast as finding corresponding points in two or more images that correspond to the same physical point in the scene [16,24,13,23,9,10,21,6].…”