Disentangling Factors of Variation by Mixing Them

Hu, Qiyang; Szabo, Attila; Portenier, Tiziano; Favaro, Paolo; Zwicker, Matthias (June 2018). Disentangling Factors of Variation by Mixing Them. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, Utah, United States. June 18 - June 22, 2018.

[img] Text
Hu_Disentangling_Factors_of_CVPR_2018_paper.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (1MB) | Request a copy

We propose an approach to learn image representations that consist of disentangled factors of variation without exploiting any manual labeling or data domain knowledge. A factor of variation corresponds to an image attribute that can be discerned consistently across a set of images, such as the pose or color of objects. Our disentangled representation consists of a concatenation of feature chunks, each chunk representing a factor of variation. It supports applications such as transferring attributes from one image to another, by simply mixing and unmixing feature chunks, and classification or retrieval based on one or several attributes, by considering a user-specified subset of feature chunks. We learn our representation without any labeling or knowledge of the data domain, using an autoencoder architecture with two novel training objectives: first, we propose an invariance objective to encourage that encoding of each attribute, and decoding of each chunk, are invariant to changes in other attributes and chunks, respectively; second, we include a classification objective, which ensures that each chunk corresponds to a consistently discernible attribute in the represented image, hence avoiding degenerate feature mappings where some chunks are completely ignored. We demonstrate the effectiveness of our approach on the MNIST, Sprites, and CelebA datasets.

Item Type:

Conference or Workshop Item (Paper)

Division/Institute:

08 Faculty of Science > Institute of Computer Science (INF)

UniBE Contributor:

Hu, Qiyang; Szabo, Attila; Portenier, Tiziano; Favaro, Paolo and Zwicker, Matthias

Subjects:

000 Computer science, knowledge & systems
500 Science > 510 Mathematics

Language:

English

Submitter:

Xiaochen Wang

Date Deposited:

28 May 2019 15:46

Last Modified:

26 Oct 2019 14:45

BORIS DOI:

10.7892/boris.126527

URI:

https://boris.unibe.ch/id/eprint/126527

Actions (login required)

Edit item Edit item
Provide Feedback