• Corpus ID: 239768926

Domain Adaptation for Rare Classes Augmented with Synthetic Samples

  title={Domain Adaptation for Rare Classes Augmented with Synthetic Samples},
  author={Tuhinanksu Das and Robert-Jan Bruintjes and Attila Lengyel and Jan C. van Gemert and Sara Beery},
To alleviate lower classification performance on rare classes in imbalanced datasets, a possible solution is to augment the underrepresented classes with synthetic samples. Domain adaptation can be incorporated in a classifier to decrease the domain discrepancy between real and synthetic samples. While domain adaptation is generally applied on completely synthetic source domains and real target domains, we explore how domain adaptation can be applied when only a single rare class is augmented… 
1 Citations

Figures from this paper

A Survey on Deep Transfer Learning and Beyond

This survey first review more than 50 representative approaches of DTL in the last decade and systematically summarize them into four categories and divides each category into subcategories according to models, functions, and operation objects.



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.

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.

Synthetic to Real Adaptation with Generative Correlation Alignment Networks

A Deep Generative Correlation Alignment Network (DGCAN) is proposed to synthesize images using a novel domain adaption algorithm that leverages a shape preserving loss and a low level statistic matching loss to minimize the domain discrepancy between synthetic and real images in deep feature space.

Domain-Adversarial Training of Neural Networks

A new representation learning approach for domain adaptation, in which data at training and test time come from similar but different distributions, which can be achieved in almost any feed-forward model by augmenting it with few standard layers and a new gradient reversal layer.

Synthetic Examples Improve Generalization for Rare Classes

The experiments reveal that synthetic data can considerably reduce error rates for classes that are rare, that as the amount of simulated data is increased, accuracy on the target class improves, and that high variation of simulation data provides maximum performance gain.

Return of Frustratingly Easy Domain Adaptation

This work proposes a simple, effective, and efficient method for unsupervised domain adaptation called CORrelation ALignment (CORAL), which minimizes domain shift by aligning the second-order statistics of source and target distributions, without requiring any target labels.

Deep CORAL: Correlation Alignment for Deep Domain Adaptation

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.

Mind the Class Weight Bias: Weighted Maximum Mean Discrepancy for Unsupervised Domain Adaptation

It is shown that MMD cannot account for class weight bias and results in degraded domain adaptation performance and a weighted MMD model is proposed, which is defined by introducing an auxiliary weight for each class in the source domain, and a classification EM algorithm is suggested.

Domain Stylization: A Strong, Simple Baseline for Synthetic to Real Image Domain Adaptation

This paper shows that by applying a straightforward modification to an existing photorealistic style transfer algorithm, it is shown that this approach exceeds the performance of any current state-of-the-art GAN-based image translation approach as measured by segmentation and object detection metrics.