IEEE Transactions on Image Processing · 2023 · 11 citations · 50 references
ComputingComputer ScienceComputer VisionFeature EngineeringMachine LearningDeep LearningGenerative ModelingRepresentation LearningCross-domain GeneralizationData AugmentationDeep Image FeaturesDomain GeneralizationSemantic Generalization
Despite remarkable success in a variety of computer vision applications, it is well-known that deep learning can fail catastrophically when presented with out-of-distribution data, where there are usually style differences between the training and test images. Toward addressing this challenge, we consider the domain generalization problem, wherein predictors are trained using data drawn from a family of related training (source) domains and then evaluated on a distinct and unseen test domain. Naively training a model on the aggregate set of data (pooled from all source domains) has been shown to perform suboptimally, since the information learned by that model might be domain-specific and generalizes imperfectly to test domains. Data augmentation has been shown to be an effective approach to overcome this problem. However, its application has been limited to enforcing invariance to simple transformations like rotation, brightness change, etc. Such perturbations do not necessarily cover plausible real-world variations that preserve the semantics of the input (such as a change in the image style). In this paper, taking the advantage of multiple source domains, we propose a novel approach to express and formalize robustness to these kind of real-world image perturbations. The three key ideas underlying our formulation are (1) leveraging disentangled representations of the images to define different factors of variations, (2) generating perturbed images by changing such factors composing the representations of the images, (3) enforcing the learner (classifier) to be invariant to such changes in the images. We use image-to-image translation models to demonstrate the efficacy of this approach. Based on this, we propose a domain-invariant regularization (DIR) loss function that enforces invariant prediction of targets (class labels) across domains which yields improved generalization performance. We demonstrate the effectiveness of our approach on several widely used datasets for the domain generalization problem, on all of which our results are competitive with the state-of-the-art.
50
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. · 2016 · 222.1K citations · Full text
An overview of statistical learning theory
Vladimir Vapnik · IEEE Transactions on Neural Networks · 1999 · 6.2K citations
Adversarial Discriminative Domain Adaptation
Eric Tzeng, Judy Hoffman, Kate Saenko et al. · 2017 · 5K citations
mixup: Beyond Empirical Risk Minimization
Hongyi Zhang, Moustapha Cissé, Yann Dauphin et al. · arXiv (Cornell University) · 2017 · 4.8K citations · Full text