Edge-preserving directional regularization technique for disparity estimation of stereoscopic images

Due to the advances in visual technology three dimensional (3D) imaging systems are becoming popular. One way of stimulating 3D perception is to use stereo pairs, a pair of images of the same scene acquired from different perspectives. Since there is an inherent redundancy between the images of a stereo pair, data compression algorithms should be employed to efficiently represent stereo pairs. Disparity estimation of stereoscopic image is an important stage for the stereo image compression. We propose a new disparity estimation method using a directional regularization technique to preserve edges well for stereoscopic image coding. This method smooths disparity vectors in smooth regions and preserves edges in object boundaries without over-smoothing problem. The experimental results show that the proposed method achieves close matches between a left image and a right image as well as the improved coding efficiency. We use DPCM technique followed by Huffman entropy coding for disparity map coding. In addition, the proposed disparity estimation method can be applied to the high-quality intermediate view reconstruction