Each dataset contains pre-rectified horizontal-parallax multi-perspective images of 3D scenes. Each dataset consists of 193 camera views (images), positioned equidistantly on a line, where the disparity range between adjacent views is 1 pixel at most. Images are of size 1280×720 pixels and are stored in 8-bit RGB format (PNG). The datasets were used in the ICME 2018 grand challenge on densely sampled light field reconstruction. Please, find the citation instructions on the dataset page.