Disentangling Factors of Variation by Mixing Them
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks2017
- Photographic Image Synthesis with Cascaded Refinement Networks2017
- Disentangled Representation Learning GAN for Pose-Invariant Face Recognition2017
- Learning from Simulated and Unsupervised Images through Adversarial Training2017
- Unsupervised Learning of Depth and Ego-Motion from Video2017
- Image-to-Image Translation with Conditional Adversarial Networks2017
- Neural Face Editing with Intrinsic Image Disentangling2017
- Context Encoders: Feature Learning by Inpainting2016
- Deep Learning Face Attributes in the Wild2015
- Auto-association by multilayer perceptrons and singular value decompositionBiological Cybernetics, 1988