TY - GEN
T1 - Deforming autoencoders
T2 - 15th European Conference on Computer Vision, ECCV 2018
AU - Shu, Zhixin
AU - Sahasrabudhe, Mihir
AU - Alp Güler, Rıza
AU - Samaras, Dimitris
AU - Paragios, Nikos
AU - Kokkinos, Iasonas
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2018.
PY - 2018
Y1 - 2018
N2 - In this work we introduce Deforming Autoencoders, a generative model for images that disentangles shape from appearance in an unsupervised manner. As in the deformable template paradigm, shape is represented as a deformation between a canonical coordinate system (‘template’) and an observed image, while appearance is modeled in deformation-invariant, template coordinates. We introduce novel techniques that allow this approach to be deployed in the setting of autoencoders and show that this method can be used for unsupervised group-wise image alignment. We show experiments with expression morphing in humans, hands, and digits, face manipulation, such as shape and appearance interpolation, as well as unsupervised landmark localization. We also achieve a more powerful form of unsupervised disentangling in template coordinates, that successfully decomposes face images into shading and albedo, allowing us to further manipulate face images.
AB - In this work we introduce Deforming Autoencoders, a generative model for images that disentangles shape from appearance in an unsupervised manner. As in the deformable template paradigm, shape is represented as a deformation between a canonical coordinate system (‘template’) and an observed image, while appearance is modeled in deformation-invariant, template coordinates. We introduce novel techniques that allow this approach to be deployed in the setting of autoencoders and show that this method can be used for unsupervised group-wise image alignment. We show experiments with expression morphing in humans, hands, and digits, face manipulation, such as shape and appearance interpolation, as well as unsupervised landmark localization. We also achieve a more powerful form of unsupervised disentangling in template coordinates, that successfully decomposes face images into shading and albedo, allowing us to further manipulate face images.
UR - https://www.scopus.com/pages/publications/85055103605
U2 - 10.1007/978-3-030-01249-6_40
DO - 10.1007/978-3-030-01249-6_40
M3 - Conference contribution
AN - SCOPUS:85055103605
SN - 9783030012489
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 664
EP - 680
BT - Computer Vision – ECCV 2018 - 15th European Conference, 2018, Proceedings
A2 - Hebert, Martial
A2 - Ferrari, Vittorio
A2 - Sminchisescu, Cristian
A2 - Weiss, Yair
PB - Springer Verlag
Y2 - 8 September 2018 through 14 September 2018
ER -