Unsupervised Domain Adaptation for Training Event-Based Networks Using Contrastive Learning and Uncorrelated Conditioning

Dayuan Jian, Mohammad Rostami; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023, pp. 18721-18731

Abstract

Event-based cameras offer reliable measurements for preforming computer vision tasks in high-dynamic range environments and during fast motion maneuvers. However, adopting deep learning in event-based vision faces the challenge of annotated data scarcity due to recency of event cameras. Transferring the knowledge that can be obtained from conventional camera annotated data offers a practical solution to this challenge. We develop an unsupervised domain adaptation algorithm for training a deep network for event-based data image classification using contrastive learning and uncorrelated conditioning of data. Our solution outperforms the existing algorithms for this purpose.

Related Material

[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Jian_2023_ICCV, author = {Jian, Dayuan and Rostami, Mohammad}, title = {Unsupervised Domain Adaptation for Training Event-Based Networks Using Contrastive Learning and Uncorrelated Conditioning}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, month = {October}, year = {2023}, pages = {18721-18731} }