CIF: Medium: Understanding Quality of 3D Video with Applications in 3D Video Processing and Communications

根据人类视觉系统中立体处理的一般原理，对3D图像/视频感知的研究正被用于为3D视频应用程序派生3D质量指标，以提供最佳的3D体验。人类观察者可以以高灵敏度探测到深度差异，但精度有限。此外，虽然视觉系统可以表示由高空间频率分量携带的二维图像的精细细节（即使当图像快速变化时），但它不能以相对高的空间或时间分辨率跟踪深度变化。因此，立体深度的表示受到带宽和位深度的限制。由于这些限制，在视网膜水平上，深度呈现的准确性的一些偏差在感知上是显著的，而另一些则不那么明显。深度信号在一系列空间和时间剖面上感知图像保真度的测量被用于指导最佳视频处理技术的开发，并允许评估替代3d视频编码算法（如多视图与视频+深度）的优点和局限性。视觉实验研究了使用各种编码方案生成的3d视频的感知保真度和感知图像质量。根据这些结果，质量指标被开发并集成到视频处理和通信应用中。一种以人为中心的视差估计和视图合成算法正在开发中，用于视频处理和通信应用；这也可以用来提高物体检测、分类和跟踪的性能，并为自动立体显示生成多视图，这在3D诊断医学成像和外科系统中得到了应用。

Research on 3D image/video perception in the light of general principles of stereoprocessing in the human visual system is being used to derive 3D quality metrics for 3Dvideo applications in order to deliver the best 3D experience.Human observers can detect differences in depth with high sensitivity, but limitedprecision. Moreover, while the visual system can represent fine details of the 2D imagethat are carried by high spatial frequency components (even when the image is rapidlychanging), it can not track variations in depth with comparably high resolution in spaceor time. Thus the representation of stereoscopic depth is restricted both in bandwidth andin bit depth. Because of those limitations, some deviations from accuracy in therepresentation of depth at the retinal level are perceptually salient, and others less so.Measurements of perceived image fidelity across a range of spatial and temporal profilesfor the depth signal are being used to guide the development of optimal video processingtechniques, and to allow evaluation of the advantages and limitations of alternative 3Dvideo coding algorithms such as multiview versus video+depth).Vision experiments investigate both perceived fidelity and perceived image quality in 3Dvideo generated using a variety of encoding schemes. From those results quality metricsare developed and integrated into video processing and communications applications. Ahuman-centric disparity estimation and view synthesis algorithm is being developed forvideo processing and communications applications; this can also be used to improve theperformance of object detection, classification and tracking, and to generate multi viewsfor autostereoscopic display, which finds applications in 3D enabled diagnostic medicalimaging and surgical systems.