Exploring Unlabeled Faces for Novel Attribute Discovery
Authors:
- 1 June 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5820-5829
Abstract
Despite remarkable success in unpaired image-to-image translation, existing systems still require a large amount of labeled images. This is a bottleneck for their real-world applications; in practice, a model trained on labeled CelebA dataset does not work well for test images from a different distribution -- greatly limiting their application to unlabeled images of a much larger quantity. In this paper, we attempt to alleviate this necessity for labeled data in the facial image translation domain. We aim to explore the degree to which you can discover novel attributes from unlabeled faces and perform high-quality translation. To this end, we use prior knowledge about the visual world as guidance to discover novel attributes and transfer them via a novel normalization method. Experiments show that our method trained on unlabeled data produces high-quality translations, preserves identity, and be perceptually realistic, as good as, or better than, state-of-the-art methods trained on labeled data.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization2017
- Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks2017
- Image-to-Image Translation with Conditional Adversarial Networks2017
- Unsupervised automatic attribute discovery method via multi-graph clustering2016
- Deep Residual Learning for Image Recognition2016
- EmotioNet: An Accurate, Real-Time Algorithm for the Automatic Annotation of a Million Facial Expressions in the Wild2016
- Deep Learning Face Attributes in the Wild2015
- RGB-D object discovery via multi-scene analysis2011
- Unbiased look at dataset bias2011
- Segmentation and Unsupervised Part-based Discovery of Repetitive Objects2010