CyCADA: Cycle-Consistent Adversarial Domain Adaptation

By Judy Hoffman, Eric Tzeng, Taesung Park, Jun-Yan Zhu, Phillip Isola, Kate Saenko, Alexei A. Efros, Trevor Darrell. 2017

3,285 citations recorded by the source corpus.

Domain adaptation is critical for success in new, unseen environments. Adversarial adaptation models applied in feature spaces discover domain invariant representations, but are difficult to visualize and sometimes fail to capture pixel-level and low-level domain shifts. Recent work has shown that generative adversarial networks combined with cycle-consistency constraints are surprisingly effective at mapping images between domains, even without the use of aligned image pairs. We propose a novel discriminatively-trained Cycle-Consistent Adversarial Domain Adaptation model. CyCADA adapts representations at both the pixel-level and feature-level, enforces cycle-consistency while leveraging a task loss, and does not require aligned pairs. Our model can be applied in a variety of visual recognition and prediction settings. We show new state-of-the-art results across multiple adaptation tasks, including digit classification and semantic segmentation of road scenes demonstrating transfer from synthetic to real world domains.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Break-It-Fix-It: Unsupervised Learning for Program Repair (2021, background): “, 2018a;b) is a common approach that uses the target-to-source model to generate noisy sources and then trains the source-to-target model to reconstruct the targets (also related to cycle-consistency (Zhu et al., 2017; Hoffman et al., 2018) and style transfer (Shen et al.”
Learning Robust Global Representations by Penalizing Local Predictive Power (2019, background): “, 2018) and coercing domain invariance via adversarial learning (or further extensions) (Ganin et al., 2016; Bousmalis et al., 2017; Tzeng et al., 2017; Xie et al., 2018; Hoffman et al., 2018; Long et al., 2018; Zhao et al., 2018b; Kumar et al., 2018; Li et al., 2018b; Zhao et al., 2018a; Schoenauer”
A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation (2016, positive): “CyCADA (Hoffman et al., 2017): The semantic consistency loss proposed in CyCADA offers a good solution to learning the mapping between more complex data distributions with relatively more semantic types.”
Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation (2022, methodology): “Due to the integration of these challenges, existing UDA methods [10,14,23,34,44,52] are incapable of handling CI-UDA.”
Unsupervised Model Adaptation for Continual Semantic Segmentation (2020, methodology): “A large group of the existing UDA algorithms for image segmentation use adversarial learning for extracting domaininvariant features (Luc et al. 2016; Bousmalis et al. 2017; Hoffman et al. 2018b; Murez et al. 2018; Saito et al. 2018; Sankaranarayanan et al. 2018; Dhouib, Redko, and Lartizien 2020).”
Adversarial Continual Learning (2020, methodology): “el to train in a fully-differentiable manner by adjusting to solve the minimax optimization problem [12]. Adversarial learning of the latent space has been extensively researched in domain adaptation [13], active learning [33], and representation learning [16,21]. While previous literature is concer”
Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment (2019, methodology): “Concerning algorithms, several prior works demonstrate empirical success of domain-adversarial approaches, (Tzeng et al., 2014; Ganin et al., 2016; Bousmalis et al., 2016; Tzeng et al.; Hoffman et al., 2017; Shu et al., 2018).”
Moment Matching for Multi-Source Domain Adaptation (2018, background): “The reconstruction is obtained via an encoder- decoder [3, 10] or a GAN discriminator, such as dualGAN [46], cycle-GAN [50], disco-GAN [16], and CyCADA [15].”
A Comprehensive Survey on Transfer Learning (2019, methodology): “[157] adopted cycle-consistency loss to ensure the structural and semantic consistency.”
Unsupervised Model Personalization While Preserving Privacy and Scalability: An Open Problem (2020, background): “Finally, deep domain adaptation introduces several un-supervised back-propagation based techniques [6, 19], with state-of-the-art introducing an adversarial loss during training [34, 9].”
Learning to Adapt Structured Output Space for Semantic Segmentation (2018, methodology): “We note that CyCADA [12] has a pixel adaptation module by transforming source domain images to the target domain and hence obtains additional training samples.”
WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020, neutral): “…domain adaptation benchmarks include training on MNIST but testing on SVHN street signs (LeCun et al., 1998; Yuval et al., 2011; Tzeng et al., 2017; Hoffman et al., 2018), as well as transfers across datasets containing different renditions (e.g., photos, clipart, sketches) in DomainNet (Peng et…”
Contrastive Adaptation Network for Unsupervised Domain Adaptation (2019, methodology): “In the absence of labeled data from the target domain, Unsupervised Domain Adaptation (UDA) methods have emerged to mitigate the domain shift in data distributions [2, 1, 5, 32, 25, 17, 3, 16].”
Sim-To-Real via Sim-To-Sim: Data-Efficient Robotic Grasping via Randomized-To-Canonical Adaptation Networks (2018, methodology): “Prior methods can be split into: (1) feature-level adaptation, where domain-invariant features are learned between source and target domains [17, 15, 57, 7, 14, 36, 5, 55], or (2) pixellevel adaptation, which focuses on re-stylizing images from the source domain to make them look like images from th”
ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation (2018, methodology): “Other approaches include self-training [51] to provide pseudo labels or generative networks to produce target data [14, 34, 43].”
Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New Outlooks (2020, methodology): “Some methods such as [278, 280] distill the knowledge from the source domain to the target domain based on adversarial learning [78] and image translation [123, 124, 200].”
Larger Norm More Transferable: An Adaptive Feature Norm Approach for Unsupervised Domain Adaptation (2018, methodology): “On par with the feature-level adversarial alignment, generative pixel-level adaptation [17, 36, 20, 4, 32] utilizes Image-to-Image translation techniques to capture the low-level domain shifts.”
Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization Without Accessing Target Domain Data (2019, methodology): “Since then, adversarial training has been utilized in most of the following works [18, 3, 41, 58] for feature alignment.”