Page 86 - 《软件学报》2024年第4期

1664 软件学报 2024 年第 35 卷第 4 期

4 结 论

本文最新提出并研究了主动学习的源域无关开集域自适应, 在不额外引入训练数据和模型参数的约束
下, 解决了严格隐私保护场景下的域差异和新的开放类别出现的问题. 本文提出了基于局部多样性选择的局
部一致性主动学习算法(LCAL), 通过挑选阈值模糊样本, 有效地促进了开放类和公共类样本的分离. LCAL对
不同类型的样本施加不同的损失, 使得模型对将主动标记样本匹配纠正后的可信开放类样本更不确定, 对可
信公共类样本更确定, 从而进一步促进这两部分样本的分离. 3 个公开基准数据集上的大量实验证明了 LCAL
的有效性, 以及主动学习的源域无关开集域自适应的可行性.

References:
[1] He KM, Zhang XY, Ren SQ, et al. Deep residual learning for image recoginition. In: Proc. of the IEEE/CVF Conf. on Computer
Vision and Pattern Recognition (CVPR). Las Vegas: IEEE Computer Society, 2016. 770−778.
[2] Huang G, Liu Z, Laurens VDM, et al. Densely connected convolutional networks. In: Proc. of the IEEE/CVF Conf. on Computer
Vision and Pattern Recognition (CVPR). Honolulu: IEEE Computer Society, 2017. 4700−4708.
[3] Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks. Communications of the
ACM, 2017, 60(6): 84−90.
[4] Torralba A, Efros AA. Unbiased look at dataset bias. In: Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recognition
(CVPR). Colorado: IEEE Computer Society, 2011. 1521−1528.
[5] Saenko K, Kulis B, Fritz M, et al. Adapting visual category models to new domains. In: Proc. of the 11th European Conf. on
Computer Vision (Computer Vision-ECCV 2010). Grete: Springer, 2010. 213−226.
[6] Inoue N, Furuta R, Yamasaki T, et al. Cross-Domain weakly-supervised object detection through progressive domain adaptation. In:
Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR). Salt Lake City: IEEE Computer Society, 2018.
5001−5009.
[7] Hsu HK, Yao CH, Tsai YH, et al. Progressive domain adaptation for object detection. In: Proc. of the Winter Conf. on Applications
of Computer Vision (WACV). Snowmass Village: IEEE, 2020. 738−746.
[8] Gopalan R, Li R, Chellappa R. Domain adaptation for object recognition: An unsupervised approach. In: Proc. of the IEEE/CVF
Int’l Conf. on Computer Vision (ICCV). Barcelona: IEEE Computer Society, 2011. 999−1006.
[9] Tsai YH, Hung WC, Schulter S, et al. Learning to adapt structured output space for semantic segmentation. In: Proc. of the IEEE/
CVF Conf. on Computer Vision and Pattern Recognition (CVPR). Salt Lake City: IEEE Computer Society, 2018. 7472−7481.
[10] Zhang YF, Niu SC, Qiu Z, et al. COVID-DA: Deep domain adaptation from typical pneumonia to COVID-19. arXiv:2005.01577,
2020.
[11] Xu GX, Liu C, Liu J, et al. Cross-site severity assessment of COVID-19 from CT images via domain adaptation. IEEE Trans. on
Medical Imaging, 2022, 41(1): 88−102.
[12] Kundu JN, Venkat N, Revanur A, et al. Towards inheritable models for open-set domain adaptation. In: Proc. of the IEEE/CVF
Conf. on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE Computer Society, 2020. 12376−12385.
[13] Liang J, Hu DP, Feng JS, et al. UMAD: Universal model adaptation under domain and category shift. arXiv:2112.08553, 2021.
[14] Luo YD, Wang ZJ, Chen ZX, et al. Source-free progressive graph learning for open-set domain adaptation. arXiv:2202.06174, 2022.
[15] Feng ZY, Xu C, Tao DC. Open-set hypothesis transfer with semantic consistency. IEEE Trans. on Image Processing, 2021, 30:
6473−6484.
[16] Su JC, Tsai YH, Sohn K, et al. Active adversarial domain adaptation. In: Proc. of the IEEE Winter Conf. on Applications of
Computer Vision (WACV). Snowmass Village: IEEE, 2020. 739−748.
[17] Fu B, Cao ZJ, Wang JM, et al. Transferable query selection for active domain adaptation. In: Proc. of the IEEE/CVF Conf. on
Computer Vision and Pattern Recognition (CVPR). Virtual: IEEE Computer Society, 2021. 7272−7281.
[18] Mathelin A, Deheeger F, Mougeot M, et al. Discrepancy-Based active learning for domain adaptation. In: Proc. of the Int’l Conf. on
Learning Representations (ICLR). Virtual: Openreview.net, 2021.
[19] Dhananjaya MM, Kumar VR, Yogamani S. Weather and light level classification for autonomous driving: Dataset, baseline and
active learning. In: Proc. of the IEEE Int’l Intelligent Transportation Systems Conf. (ITSC). Indianapolis: IEEE, 2021. 2816−2821.
[20] Riccardi G, Hakkani-Tur D. Active learning: Theory and applications to automatic speech recognition. IEEE Trans. on Speech and
Audio Processing, 2005, 13(4), 504−511.