PROJECT TITLE :

Histogram-based cost aggregation strategy with joint bilateral filtering for stereo matching

ABSTRACT:

The edge-aware bilateral filter has been demonstrated to be effective for preserving depth edges, and disparity maps obtained from Quick Bilateral Stereo (FBS) have enhanced the efficiency of algorithm and therefore the robustness to noise. However, they additionally cause a non-excellent localisation of discontinuities. To overcome this issue, a brand new bilateral filtering based mostly value aggregation utilising colour statistical classification and similarity measurement inside annular blocks is proposed during this study. We have a tendency to have adopted the similarity of histograms evaluated by Earth Mover Distance (EMD) to obtain the raw matching value within the raised annular block, since histograms are very effective and efficient in capturing the distribution characteristics of visual options. For the weights aggregation, the spatial weight is assumed to be a continuing. The color weight is calculated by using a cluster-mean-value strategy, that is implemented by the native color histogram. It improves the accuracy within the discontinuous areas. Computation redundancy is reduced by disparity candidate choice using the native minimal relevancy within the corresponding annular blocks. We have a tendency to use the potency and accuracy to demonstrate the performance of our proposed methodology. Experimental results have shown that the proposed methodology reduces the mismatch at depth discontinuous and also the computation complexity considerably.