DART: Domain-Adversarial Residual-Transfer networks for unsupervised cross-domain image classification

Fang, Xianghong; Bai, Haoli; Guo, Ziyi; Shen, Bin; Hoi, Steven C. H.; Xu, Zenglin

doi:10.1016/j.neunet.2020.03.025

Cited by 40 publications

(16 citation statements)

References 32 publications

(44 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Adversarial training Adversarial training was adopted to help learn a feature extractor that can map both the source and target input to the same feature space and let the classifier learned on the source data can transfer to the target domain (Ganin and Lempitsky 2014;Ganin et al 2016;Fang et al 2018). In detail, a classifier was adopted as the domain discriminator and the minimax optimization was implemented by using a gradient reverse layer (GRL).…”

Section: Related Workmentioning

confidence: 99%

Adversarial Training Based Multi-Source Unsupervised Domain Adaptation for Sentiment Analysis

Dai

Liu

Ren

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

Multi-source unsupervised domain adaptation (MS-UDA) for sentiment analysis (SA) aims to leverage useful information in multiple source domains to help do SA in an unlabeled target domain that has no supervised information. Existing algorithms of MS-UDA either only exploit the shared features, i.e., the domain-invariant information, or based on some weak assumption in NLP, e.g., smoothness assumption. To avoid these problems, we propose two transfer learning frameworks based on the multi-source domain adaptation methodology for SA by combining the source hypotheses to derive a good target hypothesis. The key feature of the first framework is a novel Weighting Scheme based Unsupervised Domain Adaptation framework ((WS-UDA), which combine the source classifiers to acquire pseudo labels for target instances directly. While the second framework is a Two-Stage Training based Unsupervised Domain Adaptation framework (2ST-UDA), which further exploits these pseudo labels to train a target private extractor. Importantly, the weights assigned to each source classifier are based on the relations between target instances and source domains, which measured by a discriminator through the adversarial training. Furthermore, through the same discriminator, we also fulfill the separation of shared features and private features.Experimental results on two SA datasets demonstrate the promising performance of our frameworks, which outperforms unsupervised state-of-the-art competitors.

show abstract

Section: Related Workmentioning

confidence: 99%

Adversarial Training Based Multi-Source Unsupervised Domain Adaptation for Sentiment Analysis

Dai

Liu

Ren

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

show abstract

“…Variations of generative adversarial networks (GANs) such as DTN [33], CycleGAN [34], DiscoGAN [35], UNIT [36], DART [4] have been widely used for domain adaptation of RGB images. However, not only do these methods require a large amount of data but also it is not immediately clear how to use these techniques with sparse LIDAR data nor transferring probability distributions.…”

Section: B Domain Adaptationmentioning

confidence: 99%

“…This is indeed the case in deep learning as well as in many Bayesian inference techniques. While there are many methods to adapt deep neural networks to varying domains [1], [2], [3], [4], [5], such adaptation techniques are under-explored for Bayesian models [6] despite their extensive applications in robotics [7], [8], [9], [10], [11]. As uncertainty is represented as probability distributions in Bayesian models, entire distributions need to be adapted when changing to a new domain.…”

Section: Introductionmentioning

confidence: 99%

Online Domain Adaptation for Occupancy Mapping

Tompkins¹,

Senanayake²,

Ramos³

2020

Robotics: Science and Systems XVI

View full text Add to dashboard Cite

Creating accurate spatial representations that take into account uncertainty is critical for autonomous robots to safely navigate in unstructured environments. Although recent LIDAR based mapping techniques can produce robust occupancy maps, learning the parameters of such models demand considerable computational time, discouraging them from being used in real-time and large-scale applications such as autonomous driving. Recognizing the fact that realworld structures exhibit similar geometric features across a variety of urban environments, in this paper, we argue that it is redundant to learn all geometry dependent parameters from scratch. Instead, we propose a theoretical framework building upon the theory of optimal transport to adapt model parameters to account for changes in the environment, significantly amortizing the training cost. Further, with the use of high-fidelity driving simulators and real-world datasets, we demonstrate how parameters of 2D and 3D occupancy maps can be automatically adapted to accord with local spatial changes. We validate various domain adaptation paradigms through a series of experiments, ranging from inter-domain feature transfer to simulation-to-real-world feature transfer. Experiments verified the possibility of estimating parameters with a negligible computational and memory cost, enabling large-scale probabilistic mapping in urban environments.

show abstract

“…Recently, video surveillance for traffic control and security is playing a growing influence on current public transportation systems. During the last decade, vehicle-related researches have attracted more interest and made great progress in computer vision community, such as vehicle detection [1][2], segmentation [3] [4] and classification [5] [6]. Different with the tasks above, vehicle reID aims to precisely match a certain vehicle across scenes captured from multiple nonoverlapping cameras, which plays a crucial role in constructing the smart cities [7].…”

Section: Introductionmentioning

confidence: 99%

Cross domain knowledge learning with dual-branch adversarial network for vehicle re-identification

Peng

Wang

et al. 2020

Neurocomputing

View full text Add to dashboard Cite

The widespread popularization of vehicles has facilitated all people's life during the last decades. However, the emergence of a large number of vehicles poses the critical but challenging problem of vehicle re-identification (reID). Till now, for most vehicle reID algorithms, both the training and testing processes are conducted on the same annotated datasets under supervision. However, even a well-trained model will still cause fateful performance drop due to the severe domain bias between the trained dataset and the real-world scenes.To address this problem, this paper proposes a domain adaptation framework for vehicle reID (DAVR), which narrows the cross-domain bias by fully exploiting the labeled data from the source domain to adapt the target domain. DAVR develops an image-to-image translation network named Dual-branch Adversarial Network (DAN), which could promote the images from the source domain (well-labeled) to learn the style of target domain (unlabeled) without any annotation and preserve identity information from source domain. Then the generated images are employed to train the vehicle reID model by a proposed attention-based feature learning model with more reasonable styles. Through the proposed framework, the well-trained reID model has better domain adaptation ability for various scenes in real-world situations. Comprehensive experimental results have demonstrated that our proposed DAVR can achieve excellent performances on both VehicleID dataset and VeRi-776 dataset.

show abstract

DART: Domain-Adversarial Residual-Transfer networks for unsupervised cross-domain image classification

Cited by 40 publications

References 32 publications

Adversarial Training Based Multi-Source Unsupervised Domain Adaptation for Sentiment Analysis

Adversarial Training Based Multi-Source Unsupervised Domain Adaptation for Sentiment Analysis

Online Domain Adaptation for Occupancy Mapping

Cross domain knowledge learning with dual-branch adversarial network for vehicle re-identification

Contact Info

Product

Resources

About