Contrastive Adaptation Network for Unsupervised Domain Adaptation

By Guoliang Kang, Lu Jiang, Yi Yang, Alexander G. Hauptmann. 2019

991 citations recorded by the source corpus.

Unsupervised Domain Adaptation (UDA) makes predictions for the target domain data while manual annotations are only available in the source domain. Previous methods minimize the domain discrepancy neglecting the class information, which may lead to misalignment and poor generalization performance. To address this issue, this paper proposes Contrastive Adaptation Network (CAN) optimizing a new metric which explicitly models the intra-class domain discrepancy and the inter-class domain discrepancy. We design an alternating update strategy for training CAN in an end-to-end manner. Experiments on two real-world benchmarks Office-31 and VisDA-2017 demonstrate that CAN performs favorably against the state-of-the-art methods and produces more discriminative features.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation (2022, methodology): “Step 2 [10,11,12,13,14,15,16,17,18,19] Kettle, Folder, Keyboard, Flipflops, Pencil, Bed, Hammer, ToothBrush, Couch, Bike”
Uncertainty-aware Contrastive Distillation for Incremental Semantic Segmentation (2022, background): “Self-supervised learning has recently attracted considerable attention in the computer vision community for its ability to learn discriminative features using a contrastive objective [46], [52].”
On Generalizing Beyond Domains in Cross-Domain Continual Learning (2022, background): “To name a few, the Contrastive Adaptation Network (CAN) for unsupervised domain adaptation [17], the Covariance Metric Networks (CovaMNet) for few-shot learning [22], the Model-Agnostic learning of Semantic Features (MASF) for standard domain generalization [6] and”
Continual Semantic Segmentation via Repulsion-Attraction of Sparse and Disentangled Latent Representations (2021, methodology): “also recalls features clustering (if features belong to the same class) and separation (if features belong to different classes), which has been recently applied to adapt semantic segmentation models across domains [29, 37, 61].”
A Comprehensive Survey on Transfer Learning (2019, methodology): “, the Alexnet [138], [140] pretrained on ImageNet [166] and then directly trained on the target domain) are performed on the data set Office-31 (here, we directly use the results of CDAN, JAN, CAN, and the baseline from the original articles [137], [147], [151], [158]).”
A Broader Study of Cross-Domain Few-Shot Learning (2019, neutral): “More recently, many methods have been proposed to address the problem of domain adaptation using deep neural networks, including discrepancy-based methods, designed to align marginal distributions between the domains [38,54,23,30], adversarial-based approaches, which rely on a domain discriminator t”
A Review of Single-Source Deep Unsupervised Visual Domain Adaptation (2020, background): “[46] MK-MMD domain FL shared marginal [47] CORAL domain last FL shared marginal [36] CORAL domain last (conv, FL) shared marginal [48] weight, MMD domain all linear marginal [49] CMD domain all shared marginal [50] HoMM domain FL shared marginal [51] CDD class all except BN shared marginal [69] RKHS”
Domain Consensus Clustering for Universal Domain Adaptation (2021, background): “Moreover, there are also some works performing adaptation via clustering in the target domain [28, 57].”
FSDR: Frequency Space Domain Randomization for Domain Generalization (2021, background): “Unsupervised domain adaptation (UDA) has been studied widely to tackle the domain mismatch problem by learning domain invariant/aligned features from a labelled source domain and an unlabelled target domain [22, 29, 30, 58, 66, 69, 40, 72, 5, 6].”
Harmonizing Transferability and Discriminability for Adapting Object Detectors (2020, background): “A general practice in UDA is to bridge the domain gap by explicitly learning invariant representations between domains and achieving small error on the source domain, which have achieved compelling performance on image classification [15, 55, 13, 49, 54, 45, 23, 5] and semantic segmentation [52, 19,”
A Survey of Deep Learning-Based Object Detection (2019, methodology): “er than aligning images that are globally dissimilar. When only in the source domain manual annotations are available, Unsupervised Domain Adaptation methods is to address this issue. Haupmann et al. [265] propose a Unsupervised Domain Adaptation method which models both the intra-class and the inte”
Category Anchor-Guided Unsupervised Domain Adaptation for Semantic Segmentation (2019, methodology): “To overcome this issue, a variety of domain adaptation methods for classiﬁcation [6, 23, 37, 28, 41, 3, 19], detection [38, 16], and segmentation [7, 14, 13, 27, 34, 40, 21, 47] have been proposed.”
Unleashing the Power of Contrastive Self-Supervised Visual Models via Contrast-Regularized Fine-Tuning (2021, background): “Theorem 1) and also helps to alleviate distribution shifts among domains [25].”
CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation (2021, methodology): “One kind of such UDA methods are based on category-level alignment (Kang et al., 2019; Zhang et al., 2019; Jiang et al., 2020; Li et al., 2021b), which have achieved promising results on public UDA datasets using deep convolution neural networks (CNNs).”
Contrastive Test-Time Adaptation (2022, methodology): “For UDA methods we compare to DANN [12], CDAN [32], CDAN+BSP [8], CAN [22], SWD [25], and MCC [21].”