Improving the Generalization Ability of Deep Neural Networks for Cross-Domain Visual Recognition

Feature learning with deep neural networks (DNNs) has made remarkable progress in recent years. However, its data-driven nature makes the collection of labeled training data expensive or impossible when the testing domain changes. Here, we propose a method of transferable feature learning and instan...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on cognitive and developmental systems 2021-09, Vol.13 (3), p.607-620
Main Authors: Zheng, Jianwei, Lu, Chao, Hao, Cong, Chen, Deming, Guo, Donghui
Format: Article
Language:English
Subjects:
Artificial neural networks
Cross-domain
deep neural networks (DNNs)
domain adaptation
Domains
Feature extraction
Feature recognition
generalization ability
Image classification
Machine learning
Neural networks
Object detection
Object recognition
Task analysis
Training
Training data
Video data
visual recognition
Visualization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Feature learning with deep neural networks (DNNs) has made remarkable progress in recent years. However, its data-driven nature makes the collection of labeled training data expensive or impossible when the testing domain changes. Here, we propose a method of transferable feature learning and instance-level adaptation to improve the generalization ability of DNNs so as to mitigate the domain shift challenge for cross-domain visual recognition. When less labeled information is available, our proposed method shows attractive results in the new target domain and outperforms the typical fine-tuning method. Two DNNs are chosen as the representatives working with our proposed method, to do a comprehensive study about the generalization ability on the tasks of image-to-image transfer, image-to-video transfer, multidomain image classification, and weakly supervised detection. The experimental results show that our proposed method is superior to other existing works in the literature. In addition, a large scale of cross-domain database is merged from three different domains, providing a quantitative platform to evaluate different approaches in the field of cross-domain object detection.
ISSN:2379-8920
2379-8939
DOI:10.1109/TCDS.2020.2965166