Z-Direction Motion Estimation In 2d+Depth Map Based 3d Video

Abstract

3-Dimensional (3D) Video Quality Assessment (VQA) has been an important area for today's researchers along with the growing interest in 3D video. Even though, 2-Dimensional (2D) objective VQA metrics that are widely utilized by researchers in literature, the case is not the same for the 3D video. Therefore, subjective tests, which are inefficient in terms of time and cost are used for the 3D VQA. The VQA metrics are divided into three categories according to the use of the reference signal in the assessment as Full-Reference (FR), Reduced-Reference (RR) or No-Reference (NR). The NR metrics are more efficient compared to the other metrics. In this study, an algorithm that estimates the backward-forward z direction motion of the depth maps in 2D+depth map based 3D videos is proposed. The results indicate that the proposed algorithm is efficient to develop a NR VQA metric in future.