Occlusion Robust Depth Estimation from Binocular Stereo Video
DOI:
https://doi.org/10.17762/msea.v71i4.1132Abstract
Binocular stereo vision is used for the reconstruction of depth information and it is sensitive to senses with significant occlusion. Light field imaging is an emerging computational photography method to passive depth perception by storing many angular perspectives in a single exposure. It provides a novel solution. This work investigates binocular SV and LF imaging in order to develop the binocular LF imaging system. Based on geometrical optics theory, imaging theory is created by modelling the imaging process and examining disparity properties. Using the proposed method, we can find a depth estimation for the stereo video. Video is nothing but the sequence of frames. For the video, the structure is the same as the depth map calculation applied to the image.