SOA-EPN
Scale and Orientation Aware EPI-Patch Learning for Light Field Depth Estimation
Our method base on multi-orientation EPIs learning, which performs multi-scale matching windows selection and multi-orientation EPIs fusing by training a convolutional neural network.
Refer to our previous work(ICONIP 2017): "Luo, Y., Zhou, W., Fang, J., Liang, L., Zhang, H., & Dai, G. (2017). Epi-patch based convolutional neural network for depth estimation on 4d light field".
crosshair + diagonals, 33 views
Convolutional Layers + Fully Connected Layer
Python(Tensorflow)+Matlab
Intel Core i7-4720HQ 2.60GHz + TITAN X
