5,337
edits
No edit summary |
|||
Line 45: | Line 45: | ||
==Architecture== | ==Architecture== | ||
The only neural network they use is a transforming autoencoder. | The only neural network they use is a transforming autoencoder. | ||
Details about their network are provided in the supplementary details as well as in the code. | |||
Their encoder converts images into latent points. | |||
It consists of 7 convolutional blocks which each downsample the feature map. | |||
Each block is: Conv-BatchNorm-LeakyReLU. | |||
The output of the convolutional blocks are put through a fully connected layer and reshaped into a \(200 \times 3\) matrix. | |||
Their decoder renders the latent points into a depth map from the target view. | |||
It consists of 7 blocks of: Upsample-Conv-BatchNorm-LeakyReLU. | |||
They use bilinear upsampling. | |||
==Evaluation== | ==Evaluation== | ||
==References== | ==References== |