Although various image-based domain adaptation (DA) techniques have been proposed in recent years, domain shift in videos is still not well-explored. Most previous works only evaluate performance on small-scale datasets which are saturated. Therefore, we first propose a larger-scale dataset with larger domain discrepancy: UCF-HMDB_full. Second, we investigate different DA integration methods for videos, and show that simultaneously aligning and learning temporal dynamics achieves effective alignment even without sophisticated DA methods. Finally, we propose Temporal Attentive Adversarial Adaptation Network (TA3N), which explicitly attends to the temporal dynamics using domain discrepancy for more effective domain alignment, achieving state-of-the-art performance on three video DA datasets. The code and data are released at http://github.com/cmhungsteve/TA3N.
@article{arxiv.1905.10861,
title = {Temporal Attentive Alignment for Video Domain Adaptation},
author = {Min-Hung Chen and Zsolt Kira and Ghassan AlRegib},
journal= {arXiv preprint arXiv:1905.10861},
year = {2019}
}
Comments
CVPR2019 Workshop (Learning from Unlabeled Videos). Source code: http://github.com/cmhungsteve/TA3N