Comparison with Soft3D on Kalantari et al. dataset

Here we show rendered views from Soft3D and our method on the dataset from Kalantari et al. [1]. We show the rendered image corresponding to the approximate center of the micro-lens array.

Hover over the labels at the left to select which output to view. Hold mouse down anywhere to flip to ground truth for comparison.

[1] N. K. Kalantari, T.C. Wang, and R. Ramamoorthi. Learning based view synthesis for light field cameras. In Proc. SIGGRAPH , 2016.

Ground truth
Soft3D
DeepView
DeepView (depth)
scene_Cars
scene_Flowers1
scene_Flowers2
scene_Leaves
scene_Rock
scene_Seahorse