Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation
This work addresses the challenge of representation disentanglement for visual data in scenarios lacking labeled data, offering a solution for domain adaptation, though it appears incremental as it builds on existing disentanglement and adaptation techniques.
The paper tackles the problem of learning disentangled representations without ground truth annotations by proposing a cross-domain model that transfers attribute information from annotated source data to unlabeled target data, achieving favorable performance against state-of-the-art methods in unsupervised domain adaptation tasks.
While representation learning aims to derive interpretable features for describing visual data, representation disentanglement further results in such features so that particular image attributes can be identified and manipulated. However, one cannot easily address this task without observing ground truth annotation for the training data. To address this problem, we propose a novel deep learning model of Cross-Domain Representation Disentangler (CDRD). By observing fully annotated source-domain data and unlabeled target-domain data of interest, our model bridges the information across data domains and transfers the attribute information accordingly. Thus, cross-domain joint feature disentanglement and adaptation can be jointly performed. In the experiments, we provide qualitative results to verify our disentanglement capability. Moreover, we further confirm that our model can be applied for solving classification tasks of unsupervised domain adaptation, and performs favorably against state-of-the-art image disentanglement and translation methods.