Exploiting Local Feature Patterns for Unsupervised Domain Adaptation

@article{Wen2019ExploitingLF,
  title={Exploiting Local Feature Patterns for Unsupervised Domain Adaptation},
  author={Jun Wen and Risheng Liu and Nenggan Zheng and Qian Zheng and Zhefeng Gong and Junsong Yuan},
  journal={ArXiv},
  year={2019},
  volume={abs/1811.05042}
}
Unsupervised domain adaptation methods aim to alleviate performance degradation caused by domain-shift by learning domain-invariant representations. Existing deep domain adaptation methods focus on holistic feature alignment by matching source and target holistic feature distributions, without considering local features and their multi-mode statistics. We show that the learned local feature patterns are more generic and transferable and a further local feature distribution matching enables fine… 

Figures and Tables from this paper

Hierarchical domain adaptation with local feature patterns

Norma: A Hybrid Feature Alignment for Class-Aware Unsupervised Domain Adaptation

TLDR
This paper proposes a progressive deep feature alignment, called Norma, to tackle class-aware unsupervised domain adaptation for image classification by enforcing inter-class compactness and intra-class discrepancy through a hybrid learning process.

Unsupervised Domain Adaptation via Structurally Regularized Deep Clustering

  • Hui TangKe ChenK. Jia
  • Computer Science
    2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2020
TLDR
This work describes the proposed method as Structurally Regularized Deep Clustering (SRDC), where it enhances target discrimination with clustering of intermediate network features, and enhance structural regularization with soft selection of less divergent source examples.

Discriminative Active Learning for Domain Adaptation

Deep visual unsupervised domain adaptation for classification tasks: a survey

TLDR
The survey includes the very recent papers on this topic that have not been included in the previous surveys and introduces a taxonomy by grouping methods published on unsupervised domain adaptation into groups of: discrepancy-, adversarial-, reconstruction-, representation-, and attention-based methods.

Adversarial Graph Representation Adaptation for Cross-Domain Facial Expression Recognition

TLDR
This work proposes a novel Adversarial Graph Representation Adaptation (AGRA) framework that unifies graph representation propagation with adversarial learning for cross-domain holistic-local feature co-adaptation and shows that the proposed AGRA framework achieves superior performance over previous state-of-the-art methods.

Discriminative Adversarial Domain Adaptation

TLDR
Discriminative Adversarial Domain Adaptation has a novel adversarial objective that encourages a mutually inhibitory relation between category and domain predictions for any input instance and it defines a minimax game that can promote the joint distribution alignment.

Alignment of Local and Global Features from Multiple Layers of Convolutional Neural Network for Image Classification

TLDR
The results show that local information, coming from the first layers, may significantly improve the classification performance when merged with a global descriptor extracted from a top layer of the network.

Alignment of Local and Global Features from Multiple Layers of Convolutional Neural Network for Image Classification

TLDR
The results show that local information, coming from the first layers, may significantly improve the classification performance when merged with a global descriptor extracted from a top layer of the network.

References

SHOWING 1-10 OF 39 REFERENCES

Unsupervised Visual Domain Adaptation Using Subspace Alignment

TLDR
This paper introduces a new domain adaptation algorithm where the source and target domains are represented by subspaces described by eigenvectors, and seeks a domain adaptation solution by learning a mapping function which aligns the source subspace with the target one.

Deep CORAL: Correlation Alignment for Deep Domain Adaptation

TLDR
This paper extends CORAL to learn a nonlinear transformation that aligns correlations of layer activations in deep neural networks (Deep CORAL), and shows state-of-the-art performance on standard benchmark datasets.

Domain Adaptation for Visual Applications: A Comprehensive Survey

TLDR
An overview of domain adaptation and transfer learning with a specific view on visual applications and the methods that go beyond image categorization, such as object detection or image segmentation, video analyses or learning visual attributes are overviewed.

Domain Adaptation via Transfer Component Analysis

TLDR
This work proposes a novel dimensionality reduction framework for reducing the distance between domains in a latent space for domain adaptation and proposes both unsupervised and semisupervised feature extraction approaches, which can dramatically reduce thedistance between domain distributions by projecting data onto the learned transfer components.

Learning Transferable Features with Deep Adaptation Networks

TLDR
A new Deep Adaptation Network (DAN) architecture is proposed, which generalizes deep convolutional neural network to the domain adaptation scenario and can learn transferable features with statistical guarantees, and can scale linearly by unbiased estimate of kernel embedding.

Geodesic flow kernel for unsupervised domain adaptation

TLDR
This paper proposes a new kernel-based method that takes advantage of low-dimensional structures that are intrinsic to many vision datasets, and introduces a metric that reliably measures the adaptability between a pair of source and target domains.

Multi-Adversarial Domain Adaptation

TLDR
A multi-adversarial domain adaptation (MADA) approach, which captures multimode structures to enable fine-grained alignment of different data distributions based on multiple domain discriminators and outperforms state of the art methods on standard domain adaptation datasets.

Hybrid Heterogeneous Transfer Learning through Deep Learning

TLDR
A deep learning approach to learn a feature mapping between cross-domain heterogeneous features as well as a better feature representation for mapped data to reduce the bias issue caused by the cross- domain correspondences.

Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation

TLDR
A new model called Deep Reconstruction-Classification Network (DRCN), which jointly learns a shared encoding representation for two tasks: supervised classification of labeled source data, and unsupervised reconstruction of unlabeled target data, is designed.

Adapting Visual Category Models to New Domains

TLDR
This paper introduces a method that adapts object models acquired in a particular visual domain to new imaging conditions by learning a transformation that minimizes the effect of domain-induced changes in the feature distribution.