Subspace Interpolation via Dictionary Learning for Unsupervised Domain Adaptation

By Jie Ni, Qiang Qiu, Ramalingam Chellappa. 2013

195 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Radiology, Nuclear Medicine and Imaging, Signal Processing

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 Radiology, Nuclear Medicine and Imaging

In Signal Processing

Citations in the corpus

Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, methodology): “The approach described in [64] was inspired by the manifoldbased incremental learning framework in [61].”
A Review of Single-Source Deep Unsupervised Visual Domain Adaptation (2020, methodology): “Single-source UDA methods can be divided into two categories [26, 27]: sample re-weighting [28, 29] and intermediate subspace transformation [30–34].”
Gradually Vanishing Bridge for Adversarial Domain Adaptation (2020, methodology): “In traditional UDA methods, domain discrepancy is reduced by learning transferable representations [10, 14, 28], or reweighing the importance of the samples [29].”
MetaReg: Towards Domain Generalization using Meta-Regularization (2018, methodology): “approaches to this problem include feature engineering methods [6], learning intermediate subspaces using manifolds [11][12] and dictionaries [23], etc.”
Structure-Preserved Multi-source Domain Adaptation (2016, methodology): “Typical methods include finding intermediate subspaces that describe the smooth transition from one domain to another [8], [10], [14], [23], learning common features subspace [3], [24], [27], [29], [31] or deep features [6], [28].”
Domain adaptive sparse representation-based classification (2015, methodology): “In this section, we study the convergence properties of the proposed method and briefly discuss the computational timing compared to the dictionary-based domain adaptation algorithms [16], [14].”
Domain-Adaptive Few-Shot Learning (2020, methodology): “The conventional UDA models [10, 15, 36, 13, 55, 56, 64, 29, 31] typically leverage the subspace alignment technique.”
Open Set Domain Adaptation (2017, methodology): “Jointly transforming source and target domains into a common low dimensional space was also done together with a conjugate gradient minimisation of a transformation matrix with orthogonality constraints [3] and with dictionary learning to find subspace interpolations [32, 38, 47].”
Ensemble manifold regularized sparse low-rank approximation for multiview feature embedding (2015, background): “provides more potential discriminative information to distinguish the patterns of different classes, the feature vector of each view is usually lies in a high dimensional feature space and thus combining them together frequently leads to the problem of ”curse of dimensionality” [3, 4, 5].”
Frustratingly Easy NBNN Domain Adaptation (2013, background): “As pointed out in [20, 19], a visual word dictionary built on the source set may be a bad reference for local descriptor extracted from the target.”
Cross-Modality Distillation: A Case for Conditional Generative Adversarial Networks (2018, background): “Recent work in Domain Adaptation [4, 5, 6] addresses differences between source and target domains.”
Attention Transfer from Web Images for Video Recognition (2017, background): “There are two common strategies: one approach is to reweight the instances from the source domain [4, 22], and the other approach is to find a mapping function that would align the source distribution with the target domain [1, 9, 12, 26].”
Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation (2017, methodology): “ly for the classiﬁcation task. Some of the earlier approaches involved using feature reweighting techniques [5], or constructing intermediate representations using manifolds [11] [10] or dictionaries [24]. Since the advent of deep neural networks, emphasis has been shifted to learning domain invaria”
Domain Adaptation on the Statistical Manifold (2014, background): “As a consequence, in recent years, many solutions to this domain shift problem have been proposed [26, 27, 15, 14, 13].”
Unsupervised Domain Adaptation for Zero-Shot Learning (2015, methodology): “Among them the most relevant are the unsupervised subspace alignment based approaches [13, 12, 25, 8], as sparse coding can also be considered as learning a projection to a subspace (i.”