Depth Image Based Rendering
Depth Image Based Rendering
复制标题
DOI：
10.1007/978-3-642-22407-2_12
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
Michael Schmeing;Xiaoyi Jiang
中科院分区：
文献类型：
--
作者：
Michael Schmeing;Xiaoyi Jiang
基于深度图像的渲染是一种从视频流中渲染新视图的技术。场景几何由附加的深度流给出，该深度流为每个像素存储其到相机的距离。该技术允许各种各样的应用，包括2D-3D视频转换。本章介绍了基于深度图像的渲染，并讨论了一些挑战，包括建议的解决方案。主要的焦点在于解除遮挡问题，这是由于这样一个事实，即场景只观察到从一个单一的视图，这可能会导致在生成的虚拟视图中丢失的信息。其他挑战包括不完美的深度图，重影和纸板效果。
Depth Image Based Rendering is a technique to render new views from a video stream. The scene geometry is given by an additional depth stream which stores for each pixel its distance to the camera. This technique allows for a wide variety of applications including 2D-3D video conversion. This chapter gives an introduction to Depth Image Based Rendering and discusses some challenges including proposed solutions. The main focus lies on the disocclusion problem which is due to the fact that the scene is only observed from a single view which can lead to missing information in the generated virtual views. Other challenges include imperfect depth maps, ghosting, and the cardboard effect.