Autoencoding beyond pixels using a learned similarity metric

Anders Boesen Lindbo Larsen, Søren Kaae Sønderby, Hugo Larochelle, Ole Winther
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:1558-1566, 2016.

Abstract

We present an autoencoder that leverages learned representations to better measure similarities in data space. By combining a variational autoencoder (VAE) with a generative adversarial network (GAN) we can use learned feature representations in the GAN discriminator as basis for the VAE reconstruction objective. Thereby, we replace element-wise errors with feature-wise errors to better capture the data distribution while offering invariance towards e.g. translation. We apply our method to images of faces and show that it outperforms VAEs with element-wise similarity measures in terms of visual fidelity. Moreover, we show that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic.

Cite this Paper


BibTeX
@InProceedings{pmlr-v48-larsen16, title = {Autoencoding beyond pixels using a learned similarity metric}, author = {Larsen, Anders Boesen Lindbo and Sønderby, Søren Kaae and Larochelle, Hugo and Winther, Ole}, booktitle = {Proceedings of The 33rd International Conference on Machine Learning}, pages = {1558--1566}, year = {2016}, editor = {Balcan, Maria Florina and Weinberger, Kilian Q.}, volume = {48}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {20--22 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v48/larsen16.pdf}, url = { http://proceedings.mlr.press/v48/larsen16.html }, abstract = {We present an autoencoder that leverages learned representations to better measure similarities in data space. By combining a variational autoencoder (VAE) with a generative adversarial network (GAN) we can use learned feature representations in the GAN discriminator as basis for the VAE reconstruction objective. Thereby, we replace element-wise errors with feature-wise errors to better capture the data distribution while offering invariance towards e.g. translation. We apply our method to images of faces and show that it outperforms VAEs with element-wise similarity measures in terms of visual fidelity. Moreover, we show that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic.} }
Endnote
%0 Conference Paper %T Autoencoding beyond pixels using a learned similarity metric %A Anders Boesen Lindbo Larsen %A Søren Kaae Sønderby %A Hugo Larochelle %A Ole Winther %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-larsen16 %I PMLR %P 1558--1566 %U http://proceedings.mlr.press/v48/larsen16.html %V 48 %X We present an autoencoder that leverages learned representations to better measure similarities in data space. By combining a variational autoencoder (VAE) with a generative adversarial network (GAN) we can use learned feature representations in the GAN discriminator as basis for the VAE reconstruction objective. Thereby, we replace element-wise errors with feature-wise errors to better capture the data distribution while offering invariance towards e.g. translation. We apply our method to images of faces and show that it outperforms VAEs with element-wise similarity measures in terms of visual fidelity. Moreover, we show that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic.
RIS
TY - CPAPER TI - Autoencoding beyond pixels using a learned similarity metric AU - Anders Boesen Lindbo Larsen AU - Søren Kaae Sønderby AU - Hugo Larochelle AU - Ole Winther BT - Proceedings of The 33rd International Conference on Machine Learning DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-larsen16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 48 SP - 1558 EP - 1566 L1 - http://proceedings.mlr.press/v48/larsen16.pdf UR - http://proceedings.mlr.press/v48/larsen16.html AB - We present an autoencoder that leverages learned representations to better measure similarities in data space. By combining a variational autoencoder (VAE) with a generative adversarial network (GAN) we can use learned feature representations in the GAN discriminator as basis for the VAE reconstruction objective. Thereby, we replace element-wise errors with feature-wise errors to better capture the data distribution while offering invariance towards e.g. translation. We apply our method to images of faces and show that it outperforms VAEs with element-wise similarity measures in terms of visual fidelity. Moreover, we show that the method learns an embedding in which high-level abstract visual features (e.g. wearing glasses) can be modified using simple arithmetic. ER -
APA
Larsen, A.B.L., Sønderby, S.K., Larochelle, H. & Winther, O.. (2016). Autoencoding beyond pixels using a learned similarity metric. Proceedings of The 33rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 48:1558-1566 Available from http://proceedings.mlr.press/v48/larsen16.html .

Related Material