Alternating Back-Propagation for Generator Network
Tian Han* ,
Yang Lu* ,
Song-Chun Zhu ,
and Ying Nian Wu
* Equal contributions.
University of California, Los Angeles (UCLA), USA
Main > Comparison
Recovery error
After learning the model from the training images (now assumed to be fully observed), we can evaluate the model by the reconstruction error on the testing images. We randomly select 1000 face images for training and 300 images for testing from CelebA dataset. After learning, we infer the latent factors Z for each testing image using inferential back-propagation, and then reconstruct the testing image.
| experiment | d = 20 | d = 60 | d = 100 | d = 200 |
|---|---|---|---|---|
| ABP | .0810 | .0617 | .0549 | .0523 |
| PCA | .1038 | .0820 | .0722 | .0621 |
Original images
Recovery images with d = 20
The first image canvas is reconstruct by PCA, and the second one is reconstructed by our method.
Recovery images with d = 60
The first image canvas is reconstruct by PCA, and the second one is reconstructed by our method.
Recovery images with d = 100
The first image canvas is reconstruct by PCA, and the second one is reconstructed by our method.
Recovery images with d = 200
The first image canvas is reconstruct by PCA, and the second one is reconstructed by our method.
