The i-RevNet and its dual. The inverse can be obtained from the forward model with minimal adaption and is an i-RevNet as well. Read the paper for theoretical background and detailed analysis of the trained models.
Evaluate pre-trained model on Imagenet validation set, yields 74.018% top-1 accuracy
$ bash scripts/evaluate_ilsvrc-2012.sh
Invert output of last layer on Imagenet validation set and save example images
$ bash scripts/invert_ilsvrc-2012.sh
Imagenet ILSVRC-2012 Results
i-RevNets perform on par with baseline RevNet and ResNet.
Model:
ResNet
RevNet
i-RevNet (a)
i-RevNet (b)
Val Top-1 Error:
24.7
25.2
24.7
26.0
Reconstructions from ILSVRC-2012 validation set. Top row original image, bottom row reconstruction from final representation.
Contribute
Contributions are very welcome.
Cite
@inproceedings{
jacobsen2018irevnet,title={i-RevNet: Deep Invertible Networks},author={Jörn-Henrik Jacobsen and Arnold W.M. Smeulders and Edouard Oyallon},booktitle={International Conference on Learning Representations},year={2018},url={https://openreview.net/forum?id=HJsjkMb0Z},
}