New disparity estimation scheme based on adaptive matching windows for intermediate view reconstruction

Abstract

A new intermediate view reconstruction technique based on the adaptive disparity estimation algorithm is proposed. First, the feature values indicating the local image complexity are extracted from the input stereo image pair by using edge detection and matching algorithms. Then, the matching window size for disparity estimation is adaptively selected depending on the magnitude of these feature values. That is, for the region having larger feature values, the smaller matching window size is selected, while for the opposite case, the larger matching window size is selected by comparison with a predetermined threshold value. This new approach is not only able to reduce the mismatch of the disparity vector, which occurs in conventional fine disparity estimation with a small matching window size, but also can reduce the blocking effect, which occurs in coarse disparity estimation with a large matching window size. Some experimental results show that the proposed algorithm improves the peak signal-to-noise of the reconstructed intermediate view up to 2.93 to 4.09 dB and reduces the execution time to 39.34 to 65.62% on average in comparison with those of conventional algorithms.