Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: the Benefit of Target Expectation Maximization

Abstract

In this paper, we make two contributions to unsupervised domain adaptation (UDA) using the convolutional neural network (CNN). First, our approach transfers knowledge in all the convolutional layers through attention alignment. Most previous methods align high-level representations, e.g., activations of the fully connected (FC) layers. In these methods, however, the convolutional layers which underpin critical low-level domain knowledge cannot be updated directly towards reducing domain discrepancy. Specifically, we assume that the discriminative regions in an image are relatively invariant to image style changes. Based on this assumption, we propose an attention alignment scheme on all the target convolutional layers to uncover the knowledge shared by the source domain. Second, we estimate the posterior label distribution of the unlabeled data for target network training. Previous methods, which iteratively update the pseudo labels by the target network and refine the target network by the updated pseudo labels, are vulnerable to label estimation errors. Instead, our approach uses category distribution to calculate the cross-entropy loss for training, thereby ameliorating the error accumulation of the estimated labels. The two contributions allow our approach to outperform the state-of-the-art methods by +2.6% on the Office-31 dataset.

Accepted by ECCV 2018

References in corpus (10)

• Learning Transferable Features with Deep Adaptation Networks
• Deep Domain Confusion: Maximizing for Domain Invariance
• Domain Separation Networks
• Unsupervised Image-to-Image Translation Networks
• Semantic Segmentation using Adversarial Networks
• Adversarial Discriminative Domain Adaptation
• Adversarial Complementary Learning for Weakly Supervised Object Localization
• Style Aggregated Network for Facial Landmark Detection
• Associative Domain Adaptation
• From source to target and back: symmetric bi-directional adaptive GAN

Cited by in corpus (7)

• Multi-Representation Adaptation Network for Cross-domain Image Classification
• Class Activation Map Generation by Representative Class Selection and Multi-Layer Feature Fusion
• Unsupervised Domain Adaptation: from Simulation Engine to the RealWorld
• Unsupervised Domain Adaptation using Generative Models and Self-ensembling
• Discriminative Clustering for Robust Unsupervised Domain Adaptation
• Learning Classifiers for Domain Adaptation, Zero and Few-Shot Recognition Based on Learning Latent Semantic Parts
• Specular-to-Diffuse Translation for Multi-View Reconstruction