Just DIAL: DomaIn Alignment Layers for Unsupervised Domain Adaptation
Preprint
Abstract
The empirical fact that classifiers, trained on given data collections, perform poorly when tested on data acquired in different settings is theoretically explained in domain adaptation through a shift among distributions of the source and target domains. Alleviating the domain shift problem, especially in the challenging setting where no labeled data are available for the target domain, is paramount for having visual recognition systems working in the wild. As the problem stems from a shift among distributions, intuitively one should try to align them. In the literature, this has resulted in a stream of works attempting to align the feature representations learned from the source and target domains. Here we take a different route. Rather than introducing regularization terms aiming to promote the alignment of the two representations, we act at the distribution level through the introduction of \emph{DomaIn Alignment Layers} (\DIAL), able to match the observed source and target data distributions to a reference one. Thorough experiments on three different public benchmarks we confirm the power of our approach.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Learning the Roots of Visual Domain Shift2016
- Deep Residual Learning for Image Recognition2016
- A Testbed for Cross-Dataset Analysis2015
- Deep Learning of Scene-Specific Classifier for Pedestrian Detection2014
- Unsupervised Visual Domain Adaptation Using Subspace Alignment2013
- Selective Transfer Machine for Personalized Facial Action Unit Detection2013
- Transfer Sparse Coding for Robust Image Representation2013
- Geodesic flow kernel for unsupervised domain adaptation2012
- No Bias Left behind: Covariate Shift Adaptation for Discriminative 3D Pose Estimation2012
- Domain Adaptation via Transfer Component AnalysisIEEE Transactions on Neural Networks, 2010