Abstract
We propose a method for semantic segmentation in the unsupervised domain adaptation (UDA) setting. We particularly examine the domain gap between spatial-class distributions and propose to align the local distributions of the segmentation predictions. Despite its simplicity, the proposed method achieves state-of-the-art results in UDA segmentation benchmarks.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We denote label and prediction corresponding to a pixel coordinates of (i, j) with \(y_{ij} \in \{0,1\}^{K}\) and \(f(x)_{ij} \in \mathbb {R}^K\) respectively.
- 2.
Note that this is not exactly entropy of the predictions as the \(y_{kij}\) terms are not summed over the class dimension k. But, the source sample predictions have low entropy as cross-entropy is minimized on them. Adversarial alignment results in aligned \(\overline{y}_{kij}\) distributions and thus, results in low entropy for the target predictions as well. As in [8], we found this weighted-scheme more effective because this adversarial loss promotes both the low entropy target predictions and aligned prediction maps across domains.
References
Wang, M., Deng, W.: Deep visual domain adaptation: A survey. Neurocomputing (2018)
Sun, B., Feng, J., Saenko, K.: Return of frustratingly easy domain adaptation. In: AAAI, vol. 6 (2016)
Tzeng, E., Hoffman, J., Saenko, K., Darrell, T.: Adversarial discriminative domain adaptation. In: Computer Vision and Pattern Recognition (CVPR), vol. 1 (2017)
Cicek, S., Soatto, S.: Unsupervised domain adaptation via regularized conditional alignment. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1416–1425 (2019)
Cicek, S., Fawzi, A., Soatto, S.: SaaS: speed as a supervisor for semi-supervised learning. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018, Part II. LNCS, vol. 11206, pp. 152–166. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01216-8_10
Cicek, S., Soatto, S.: Input and weight space smoothing for semi-supervised learning. In: Proceedings of the IEEE International Conference on Computer Vision Workshops (2019)
Cordts, M., et al.: The cityscapes dataset for semantic urban scene understanding. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3213–3223 (2016)
Vu, T.H., Jain, H., Bucher, M., Cord, M., Pérez, P.: Advent: Adversarial entropy minimization for domain adaptation in semantic segmentation. arXiv preprint arXiv:1811.12833 (2018)
Hoffman, J., Wang, D., Yu, F., Darrell, T.: Fcns in the wild: Pixel-level adversarial and constraint-based adaptation. arXiv preprint arXiv:1612.02649 (2016)
Chen, Y.H., Chen, W.Y., Chen, Y.T., Tsai, B.C., Frank Wang, Y.C., Sun, M.: No more discrimination: Cross city adaptation of road scene segmenters. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1992–2001 (2017)
Zhang, Y., David, P., Gong, B.: Curriculum domain adaptation for semantic segmentation of urban scenes. In: The IEEE International Conference on Computer Vision (ICCV), vol. 2 (2017)
Isola, P., Zhu, J.Y., Zhou, T., Efros, A.A.: Image-to-image translation with conditional adversarial networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1125–1134 (2017)
Tsai, Y.H., Hung, W.C., Schulter, S., Sohn, K., Yang, M.H., Chandraker, M.: Learning to adapt structured output space for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7472–7481 (2018)
Chen, Y., Li, W., Van Gool, L.: Road: reality oriented adaptation for semantic segmentation of urban scenes. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7892–7901 (2018)
Saito, K., Watanabe, K., Ushiku, Y., Harada, T.: Maximum classifier discrepancy for unsupervised domain adaptation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3723–3732 (2018)
Luo, Y., Zheng, L., Guan, T., Yu, J., Yang, Y.: Taking a closer look at domain shift: category-level adversaries for semantics consistent domain adaptation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2507–2516 (2019)
Zou, Y., Yu, Z., Kumar, B., Wang, J.: Domain adaptation for semantic segmentation via class-balanced self-training. arXiv preprint arXiv:1810.07911 (2018)
Zou, Y., Yu, Z., Liu, X., Kumar, B., Wang, J.: Confidence regularized self-training. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5982–5991 (2019)
Zou, Y., Yu, Z., Vijaya Kumar, B.V.K., Wang, J.: Unsupervised domain adaptation for semantic segmentation via class-balanced self-training. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018, Part III. LNCS, vol. 11207, pp. 297–313. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01219-9_18
Lian, Q., Lv, F., Duan, L., Gong, B.: Constructing self-motivated pyramid curriculums for cross-domain semantic segmentation: a non-adversarial approach. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 6758–6767 (2019)
Ros, G., Sellart, L., Materzynska, J., Vazquez, D., Lopez, A.M.: The synthia dataset: a large collection of synthetic images for semantic segmentation of urban scenes. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3234–3243 (2016)
Hoffman, J., et al.: Cycada: Cycle-consistent adversarial domain adaptation. arXiv preprint arXiv:1711.03213 (2017)
Zhang, Y., Qiu, Z., Yao, T., Liu, D., Mei, T.: Fully convolutional adaptation networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 6810–6818 (2018)
Chen, L.C., Papandreou, G., Kokkinos, I., Murphy, K., Yuille, A.L.: Deeplab: semantic image segmentation with deep convolutional nets, Atrous convolution, and fully connected CRFs. IEEE Trans. Pattern Anal. Mach. Intell. 40, 834–848 (2017)
Krähenbühl, P., Koltun, V.: Parameter learning and convergent inference for dense random fields. In: International Conference on Machine Learning, pp. 513–521 (2013)
Chen, L.C., Papandreou, G., Schroff, F., Adam, H.: Rethinking atrous convolution for semantic image segmentation. arXiv preprint arXiv:1706.05587 (2017)
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., Adam, H.: Encoder-decoder with Atrous separable convolution for semantic image segmentation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018, Part VII. LNCS, vol. 11211, pp. 833–851. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01234-2_49
Zhang, Q., Zhang, J., Liu, W., Tao, D.: Category anchor-guided unsupervised domain adaptation for semantic segmentation. In: Advances in Neural Information Processing Systems, pp. 433–443 (2019)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Acknowledgment
Research supported by ONR N00014-19-1-2229.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Cicek, S., Xu, N., Wang, Z., Jin, H., Soatto, S. (2021). Spatial Class Distribution Shift in Unsupervised Domain Adaptation: Local Alignment Comes to Rescue. In: Ishikawa, H., Liu, CL., Pajdla, T., Shi, J. (eds) Computer Vision – ACCV 2020. ACCV 2020. Lecture Notes in Computer Science(), vol 12624. Springer, Cham. https://doi.org/10.1007/978-3-030-69535-4_38
Download citation
DOI: https://doi.org/10.1007/978-3-030-69535-4_38
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-69534-7
Online ISBN: 978-3-030-69535-4
eBook Packages: Computer ScienceComputer Science (R0)