Abstract
The recent development of light field cameras has received growing interest, as their rich angular information has potential benefits for many computer vision tasks. In this paper, we introduce a novel method to obtain a dense disparity map by use of ground control points (GCPs) in the light field. Previous work optimizes the disparity map by local estimation which includes both reliable points and unreliable points. To reduce the negative effect of the unreliable points, we predict the disparity at non-GCPs from GCPs. Our method performs more robustly in shadow areas than previous methods based on GCP work, since we combine color information and local disparity. Experiments and comparisons on a public dataset demonstrate the effectiveness of our proposed method.
Highlights
With the rapid development of computational photography, many computational imaging devices have been invented, based on, e.g., coded apertures [1], focal sweep [2], and light fields [3, 4], examples of the latter being the Lytro and Raytrix cameras
We focus on accurate disparity estimation using the light field
It is worth noting that the structure tensor performs poorly in the near field, which corresponds to the low-slope area in the epipolar plane image (EPI); in other words, the disparity between two views is larger than 2 pixels in these areas
Summary
With the rapid development of computational photography, many computational imaging devices have been invented, based on, e.g., coded apertures [1], focal sweep [2], and light fields [3, 4], examples of the latter being the Lytro (https://lytro.com) and Raytrix (http://www. raytrix.de) cameras. We focus on accurate disparity estimation using the light field. Unlike the wide baseline used in traditional multi-view stereo, the multi-view representation synthesized by the light field has a narrow baseline, which can provide more accurate sub-pixel disparity estimation. Previous work [13, 14] has computed an optimized disparity map based on local estimation, which includes both reliable and unreliable information. To reduce the negative effects of unreliable points, we propose to obtain a dense disparity map from certain reliable estimation points in the light field called ground control points (GCPs) [15].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
