Taking a Closer Look at Domain Shift: Category-Level Adversaries for Semantics Consistent Domain Adaptation

@article{Luo2019TakingAC,
  title={Taking a Closer Look at Domain Shift: Category-Level Adversaries for Semantics Consistent Domain Adaptation},
  author={Yawei Luo and Liang Zheng and Tao Guan and Junqing Yu and Yi Yang},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2019},
  pages={2502-2511}
}
  • Yawei Luo, Liang Zheng, Yi Yang
  • Published 25 September 2018
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
We consider the problem of unsupervised domain adaptation in semantic segmentation. [] Key Method Specifically, we reduce the weight of the adversarial loss for category-level aligned features while increasing the adversarial force for those poorly aligned. In this process, we decide how well a feature is category-level aligned between source and target by a co-training approach. In two domain adaptation tasks, i.e., GTA5 -> Cityscapes and SYNTHIA -> Cityscapes, we validate that the proposed method matches…

Figures and Tables from this paper

Category-Level Adversarial Adaptation for Semantic Segmentation Using Purified Features

The designed network, CLAN, can not only disentangle the noisy factors and suppress their influences for target tasks but also utilize those purified features to conduct a more delicate level domain calibration simultaneously, i.e., global marginal distribution and local joint distribution alignment simultaneously.

An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation

This work proposes to explicitly train a domain-invariant classifier by generating and defensing against pointwise feature space adversarial perturbations and achieves the state-of-the-art performance on two challenging domain adaptation tasks for semantic segmentation.

Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation

This work proposes a fine-grained adversarial learning strategy for class-level feature alignment while preserving the internal structure of semantics across domains, which outperforms other global feature alignment based and class-wise alignment based counterparts.

Contextual-Relation Consistent Domain Adaptation for Semantic Segmentation

The proposed CrCDA learns and enforces the prototypical local contextual-relations explicitly in the feature space of a labelled source domain while transferring them to an unlabelled target domain via backpropagation-based adversarial learning.

MLAN: Multi-Level Adversarial Network for Domain Adaptive Semantic Segmentation

Adversarial Domain Adaptation with Semantic Consistency for Cross-Domain Image Classification

This paper introduces an Adaptive Centroid Alignment strategy and a Class Discriminative Constraint step to complement each other iteratively and alternatively in an end-to-end framework and shows that ADASC outperforms the state-of-the-art methods.

Significance-Aware Information Bottleneck for Domain Adaptive Semantic Segmentation

This work equips the adversarial network with a "significance-aware information bottleneck (SIB)", to address the above problem, and shows that the proposed method can yield leading results compared with other feature-space alternatives.

Semantic Distribution-aware Contrastive Adaptation for Semantic Segmentation

A new semantic distribution-aware contrastive adaptation algorithm, dubbed as SDCA, that enables pixel-wise representation alignment across domains under the guidance of the semantic distributions, and empirically unveil the certain mechanisms that promote the potential of SDCA.

Joint Adversarial Learning for Domain Adaptation in Semantic Segmentation

This work proposes a novel joint adversarial learning (JAL) to boost the domain discriminator in output space by introducing the information ofdomain discriminator from low-level features and proposes a weight transfer module (WTM) to alleviate the inherent bias of the trained decoder towards source domain.

Category-Adaptive Domain Adaptation for Semantic Segmentation

This work introduces adversarial learning into style gap bridging mechanism to keep the style information from two domains in a similar space and proposes a category-adaptive threshold mechanism to choose category-wise pseudo labels for self-supervised learning (SSL).
...

References

SHOWING 1-10 OF 55 REFERENCES

Significance-Aware Information Bottleneck for Domain Adaptive Semantic Segmentation

This work equips the adversarial network with a "significance-aware information bottleneck (SIB)", to address the above problem, and shows that the proposed method can yield leading results compared with other feature-space alternatives.

Unsupervised Domain Adaptation for Semantic Segmentation with GANs

This work proposes an approach based on Generative Adversarial Networks (GANs) that brings the embeddings closer in the learned feature space and can achieve state of the art results on two challenging scenarios of synthetic to real domain adaptation.

Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation

This work proposes an approach based on Generative Adversarial Networks (GANs) that brings the embeddings closer in the learned feature space and can achieve state of the art results on two challenging scenarios of synthetic to real domain adaptation.

FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation

This paper introduces the first domain adaptive semantic segmentation method, proposing an unsupervised adversarial approach to pixel prediction problems, and outperforms baselines across different settings on multiple large-scale datasets.

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

This paper proposes an attention alignment scheme on all the target convolutional layers to uncover the knowledge shared by the source domain, and estimates the posterior label distribution of the unlabeled data for target network training.

Learning to Adapt Structured Output Space for Semantic Segmentation

A multi-level adversarial network is constructed to effectively perform output space domain adaptation at different feature levels and it is shown that the proposed method performs favorably against the state-of-the-art methods in terms of accuracy and visual quality.

CyCADA: Cycle-Consistent Adversarial Domain Adaptation

A novel discriminatively-trained Cycle-Consistent Adversarial Domain Adaptation model that adapts representations at both the pixel-level and feature-level, enforces cycle-consistency while leveraging a task loss, and does not require aligned pairs is proposed.

Macro-Micro Adversarial Network for Human Parsing

Compared with traditional adversarial networks, the proposed Macro-Micro Adversarial Net not only enforces local and semantic consistency explicitly, but also avoids the poor convergence problem of adversarial Networks when handling high resolution images.

Adversarial Discriminative Domain Adaptation

It is shown that ADDA is more effective yet considerably simpler than competing domain-adversarial methods, and the promise of the approach is demonstrated by exceeding state-of-the-art unsupervised adaptation results on standard domain adaptation tasks as well as a difficult cross-modality object classification task.

Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks

This generative adversarial network (GAN)-based method adapts source-domain images to appear as if drawn from the target domain, and outperforms the state-of-the-art on a number of unsupervised domain adaptation scenarios by large margins.
...