Reusing the Task-specific Classifier as a Discriminator: Discriminator-free Adversarial Domain Adaptation

Chen, Lin; Chen, Huaian; Wei, Zhixiang; Jin, Xian-Min; Tan, Xiao; Jin, Yi; Chen, Enhong

doi:10.1109/cvpr52688.2022.00704

Cited by 114 publications

(34 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…target) distributions. Several research fields attempt to address this problem, such as unsupervised domain adaptation [63,6,53,56,46,58] and domain generalization [75,8]. In particular, domain generalization aims to learn invariant representation so as to cover the possible shifts of test data.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularization

Song¹,

Lee²,

Kweon³

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper presents a simple yet effective approach that improves continual test-time adaptation (TTA) in a memoryefficient manner. TTA may primarily be conducted on edge devices with limited memory, so reducing memory is crucial but has been overlooked in previous TTA studies. In addition, long-term adaptation often leads to catastrophic forgetting and error accumulation, which hinders applying TTA in real-world deployments. Our approach consists of two components to address these issues. First, we present lightweight meta networks that can adapt the frozen original networks to the target domain. This novel architecture minimizes memory consumption by decreasing the size of intermediate activations required for backpropagation. Second, our novel self-distilled regularization controls the output of the meta networks not to deviate significantly from the output of the frozen original networks, thereby preserving well-trained knowledge from the source domain. Without additional memory, this regularization prevents error accumulation and catastrophic forgetting, resulting in stable performance even in long-term test-time adaptation. We demonstrate that our simple yet effective strategy outperforms other state-of-the-art methods on various benchmarks for image classification and semantic segmentation tasks. Notably, our proposed method with ResNet-50 and WideResNet-40 takes 86% and 80% less memory than the recent state-of-the-art method, CoTTA.

show abstract

Section: Related Workmentioning

confidence: 99%

“…We refer to the baselines for which the code was officially released: TENT 3 , TTT++ 4 , CoTTA 5 , EATA 6 , and NOTE 7 . We did experiments on their code by adding the needed data loader or pre-trained model loader.…”

Section: E Baseline Details E1 Tta Workmentioning

confidence: 99%

EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularization

Song¹,

Lee²,

Kweon³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Despite aligning PLD via an adversarial learning manner can achieve effective adaptation. However, only considering intra-class-wise while neglecting inter-class-wise discrepancy may generate inaccurate results [14]. Thus, to further adaptation performance, we add inter-class-wise alignment relying on the prediction matrics from the classifier as domain critic.…”

Section: Nuclear Norm-based Discrepancymentioning

confidence: 99%

“…In adversarialbased methods, a domain discriminator is designed to encourage domain-level feature alignment via an adversarial minmax two-player game. Encouraged by the remarkable performance achieved by adversarial learning [14], we developed our method based on the adversarial paradigm.…”

Section: Introductionmentioning

confidence: 99%

“…To alleviate these problems, inspired by the finding that the implicit discriminative ability of the original classifier [14], and the localization ability of convolutional neural networks shown in [20], we propose Crucial Semantic Classifier-based Adversarial Learning for UDA. It is noted that, in terms of network architecture, we have no additional domain discriminator design.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Crucial Semantic Classifier-based Adversarial Learning for Unsupervised Domain Adaptation

Gao¹,

Li²,

Yin³

et al. 2023

Preprint

View full text Add to dashboard Cite

Unsupervised Domain Adaptation (UDA), which aims to explore the transferrable features from a well-labeled source domain to a related unlabeled target domain, has been widely progressed. Nevertheless, as one of the mainstream, existing adversarial-based methods neglect to filter the irrelevant semantic knowledge, hindering adaptation performance improvement. Besides, they require an additional domain discriminator that strives extractor to generate confused representations, but discrete designing may cause model collapse. To tackle the above issues, we propose Crucial Semantic Classifier-based Adversarial Learning (CSCAL), which pays more attention to crucial semantic knowledge transferring and leverages the classifier to implicitly play the role of domain discriminator without extra network designing. Specifically, in intra-class-wise alignment, a Paired-Level Discrepancy (PLD) is designed to transfer crucial semantic knowledge. Additionally, based on classifier predictions, a Nuclear Norm-based Discrepancy (NND) is formed that considers inter-class-wise information and improves the adaptation performance. Moreover, CSCAL can be effortlessly merged into different UDA methods as a regularizer and dramatically promote their performance.

show abstract

Fusing feature and output space for unsupervised domain adaptation on medical image segmentation

Wang

et al. 2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Image segmentation requires large amounts of annotated data. However, collecting massive datasets with annotations is difficult since they are expensive and labor‐intensive. The unsupervised domain adaptation (UDA) for image segmentation is a promising approach to address the label‐scare problem on the target domain, which enables the trained model on the source labeled domain to be adaptive to the target domain. The adversarial‐based methods encourage extracting the domain‐invariant features by training a domain discriminator to mitigate the domain gap. Existing UDA segmentation methods fail to obtain satisfied segmentation results as they only consider the global knowledge of output space while neglecting the local information of feature space. In this paper, a fusing feature and output (FFO) space method is proposed for UDA, which in the context of medical image segmentation. The proposed model is learned by training a more powerful domain discriminator, which considers features extracted from both feature space and output space. Extensive experiments carried out on several medical image datasets show the adaptation effectiveness of our approach in improving the segmentation performance.

show abstract

Reusing the Task-specific Classifier as a Discriminator: Discriminator-free Adversarial Domain Adaptation

Cited by 114 publications

References 20 publications

EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularization

EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularization

Crucial Semantic Classifier-based Adversarial Learning for Unsupervised Domain Adaptation

Fusing feature and output space for unsupervised domain adaptation on medical image segmentation

Contact Info

Product

Resources

About