Super-Resolution in Plenoptic Cameras by the Integration of Depth from Focus and Stereo

Depth range cameras like the plenoptic camera are a promising solution for 3DTV. The use of plenoptic cameras for 3DTV has some benefits with respect to 3D capture systems based on camera arrays since there is no need of geometric and color calibration or frame synchronization. This paper presents a new technique to simultaneously recover the 3D depth and all-in-focus images from a plenoptic camera integrating depth from focus and stereo methods. The standard approach to 3D reconstruction using plenoptic cameras presented some drawbacks since its spatio-angular sampling leads to a low spatial resolution. This new technique theoretically increases the spatial resolution of plenoptic cameras to its full sensor size. From a practical point of view, in typical scenes we are able to increase 49 times the spatial resolution of the standard approach and 2 times the spatial resolution of previous super-resolution techniques. Depth resolution is also increased 2 times. The resolution obtained can be used as the input of some autostereoscopic displays on the market and shows the feasibility of a plenoptic approach to 3DTV.