Cross-domain Activity Recognition via Substructural Optimal Transport

@article{Lu2021CrossdomainAR,
  title={Cross-domain Activity Recognition via Substructural Optimal Transport},
  author={Wang Lu and Yiqiang Chen and Jindong Wang and Xin Qin},
  journal={Neurocomputing},
  year={2021},
  volume={454},
  pages={65-75}
}
Abstract It is expensive and time-consuming to collect sufficient labeled data for human activity recognition (HAR). Domain adaptation is a promising approach for cross-domain activity recognition. Existing methods mainly focus on adapting cross-domain representations via domain-level, class-level, or sample-level distribution matching. However, they might fail to capture the fine-grained locality information in activity data. The domain- and class-level matching are too coarse that may result… Expand

References

SHOWING 1-10 OF 72 REFERENCES
Stratified Transfer Learning for Cross-domain Activity Recognition
TLDR
The proposed framework, referred to as Stratified Transfer Learning (STL), can dramatically improve the classification accuracy for cross-domain activity recognition and extensively investigate the performance of STL across different degrees of similarities and activity levels between domains. Expand
Cross-Dataset Activity Recognition via Adaptive Spatial-Temporal Transfer Learning
TLDR
An Adaptive Spatial-Temporal Transfer Learning (ASTTL) approach to tackle both of the above two challenges in cross-dataset HAR and can be used for both source domain selection and accurate activity transfer. Expand
Local Domain Adaptation for Cross-Domain Activity Recognition
TLDR
Experimental results on two public HAR benchmark datasets show that LDA outperforms state-of-the-art DA methods for the cross-domain HAR, and is superior to the class-to-class alignment because it can provide more accurate soft labels for the target domain. Expand
Unsupervised Domain Adaptation Using Regularized Hyper-Graph Matching
  • Debasmit Das, C. Lee
  • Computer Science, Mathematics
  • 2018 25th IEEE International Conference on Image Processing (ICIP)
  • 2018
TLDR
This work has developed a computationally efficient algorithm by initially selecting a subset of the samples to construct a graph and then developing a customized optimization routine for graph-matching based on Conditional Gradient and Alternating Direction Multiplier Method, which allows the proposed method to be used widely. Expand
Optimal Transport for Domain Adaptation
TLDR
A regularized unsupervised optimal transportation model to perform the alignment of the representations in the source and target domains, that consistently outperforms state of the art approaches and can be easily adapted to the semi-supervised case where few labeled samples are available in the target domain. Expand
Scaling Human Activity Recognition via Deep Learning-based Domain Adaptation
TLDR
A transductive transfer learning model that is specifically tuned to the properties of convolutional neural networks (CNNs) is proposed, called HDCNN, which assumes that the relative distribution of weights in the different CNN layers will remain invariant, as long as the set of activities being monitored does not change. Expand
Cross-Domain Recognition by Identifying Joint Subspaces of Source Domain and Target Domain.
TLDR
The proposed method circumvents the limitation of the global domain shift, and solves the cross-domain recognition by finding the joint subspaces of the source and target domains. Expand
Deep Subdomain Adaptation Network for Image Classification
TLDR
This work presents a deep subdomain adaptation network (DSAN) that learns a transfer network by aligning the relevant subdomain distributions of domain-specific layer activations across different domains based on a local maximum mean discrepancy (LMMD). Expand
Sample-to-Sample Correspondence for Unsupervised Domain Adaptation
TLDR
An unsupervised version of domain adaptation that considers the presence of only unlabelled data in the target domain is proposed, which out-performs traditional moment-matching methods and is competitive with respect to current local domain-adaptation methods. Expand
Transfer learning for activity recognition: a survey
TLDR
The literature is surveyed to highlight recent advances in transfer learning for activity recognition, and existing approaches to transfer-based activity recognition are characterized by sensor modality, by differences between source and target environments, by data availability, and by type of information that is transferred. Expand
...
1
2
3
4
5
...