Advertisement

Shuffle and Attend: Video Domain Adaptation

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12357)

Abstract

We address the problem of domain adaptation in videos for the task of human action recognition. Inspired by image-based domain adaptation, we can perform video adaptation by aligning the features of frames or clips of source and target videos. However, equally aligning all clips is sub-optimal as not all clips are informative for the task. As the first novelty, we propose an attention mechanism which focuses on more discriminative clips and directly optimizes for video-level (cf. clip-level) alignment. As the backgrounds are often very different between source and target, the source background-corrupted model adapts poorly to target domain videos. To alleviate this, as a second novelty, we propose to use the clip order prediction as an auxiliary task. The clip order prediction loss, when combined with domain adversarial loss, encourages learning of representations which focus on the humans and objects involved in the actions, rather than the uninformative and widely differing (between source and target) backgrounds. We empirically show that both components contribute positively towards adaptation performance. We report state-of-the-art performances on two out of three challenging public benchmarks, two based on the UCF and HMDB datasets, and one on Kinetics to NEC-Drone datasets. We also support the intuitions and the results with qualitative results.

Notes

Acknowledgment

This work was supported in part by NSF under Grant No. 1755785 and a Google Faculty Research Award. We thank NVIDIA Corporation for the GPU donation.

Supplementary material

504453_1_En_40_MOESM1_ESM.pdf (827 kb)
Supplementary material 1 (pdf 827 KB)

References

  1. 1.
    Cai, Q., Pan, Y., Ngo, C.W., Tian, X., Duan, L., Yao, T.: Exploring object relation in mean teacher for cross-domain detection. In: CVPR (2019)Google Scholar
  2. 2.
    Carreira, J., Zisserman, A.: Quo vadis, action recognition? A new model and the kinetics dataset. In: CVPR (2017)Google Scholar
  3. 3.
    Chen, M.H., Kira, Z., AlRegib, G., Woo, J., Chen, R., Zheng, J.: Temporal attentive alignment for large-scale video domain adaptation. In: ICCV (2019)Google Scholar
  4. 4.
    Chen, M.H., Li, B., Bao, Y., AlRegib, G., Kira, Z.: Action segmentation with joint self-supervised temporal domain adaptation. In: CVPR (2020)Google Scholar
  5. 5.
    Chen, M., Xue, H., Cai, D.: Domain adaptation for semantic segmentation with maximum squares loss. In: ICCV (2019)Google Scholar
  6. 6.
    Chen, Y., Li, W., Sakaridis, C., Dai, D., Van Gool, L.: Domain adaptive faster r-cnn for object detection in the wild. In: CVPR (2018)Google Scholar
  7. 7.
    Chen, Y.C., Lin, Y.Y., Yang, M.H., Huang, J.B.: Crdoco: pixel-level domain transfer with cross-domain consistency. In: CVPR (2019)Google Scholar
  8. 8.
    Cho, K., et al.: Learning phrase representations using rnn encoder-decoder for statistical machine translation. In: EMNLP (2014)Google Scholar
  9. 9.
    Choi, J., Gao, C., Messou, J.C., Huang, J.B.: Why can’t i dance in the mall? learning to mitigate scene bias in action recognition. In: NeurIPS (2019)Google Scholar
  10. 10.
    Choi, J., Sharma, G., Chandraker, M., Huang, J.B.: Unsupervised and semi-supervised domain adaptation for action recognition from drones. In: WACV (2020)Google Scholar
  11. 11.
    Feichtenhofer, C., Fan, H., Malik, J., He, K.: Slowfast networks for video recognition. In: ICCV (2019)Google Scholar
  12. 12.
    Gaidon, A., Harchaoui, Z., Schmid, C.: Temporal localization of actions with actoms. TPAMI 35(11), 2782–2795 (2013)CrossRefGoogle Scholar
  13. 13.
    Ganin, Y., Lempitsky, V.: Unsupervised domain adaptation by backpropagation. In: ICML (2015)Google Scholar
  14. 14.
    Gao, C., Zou, Y., Huang, J.B.: ican: instance-centric attention network for human-object interaction detection. In: BMVC (2018)Google Scholar
  15. 15.
    Gidaris, S., Singh, P., Komodakis, N.: Unsupervised representation learning by predicting image rotations. In: ICLR (2018)Google Scholar
  16. 16.
    Girdhar, R., Ramanan, D.: Attentional pooling for action recognition. In: NeurIPS (2017)Google Scholar
  17. 17.
    He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask R-CNN. In: ICCV (2017)Google Scholar
  18. 18.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)Google Scholar
  19. 19.
    He, Z., Zhang, L.: Multi-adversarial faster-rcnn for unrestricted object detection. In: ICCV (2019)Google Scholar
  20. 20.
    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)
  21. 21.
    Hsu, H.K., et al.: Progressive domain adaptation for object detection. In: WACV (2020)Google Scholar
  22. 22.
    Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: CVPR (2017)Google Scholar
  23. 23.
    Jamal, A., Namboodiri, V.P., Deodhare, D., Venkatesh, K.: Deep domain adaptation in action space. In: BMVC (2018)Google Scholar
  24. 24.
    Jetley, S., Lord, N.A., Lee, N., Torr, P.H.: Learn to pay attention. In: ICLR (2018)Google Scholar
  25. 25.
    Kar, A., Rai, N., Sikka, K., Sharma, G.: Adascan: adaptive scan pooling in deep convolutional neural networks for human action recognition in videos. In: CVPR (2017)Google Scholar
  26. 26.
    Khodabandeh, M., Vahdat, A., Ranjbar, M., Macready, W.G.: A robust learning approach to domain adaptive object detection. In: ICCV (2019)Google Scholar
  27. 27.
    Korbar, B., Tran, D., Torresani, L.: Scsampler: sampling salient clips from video for efficient action recognition. In: ICCV (2019)Google Scholar
  28. 28.
    Kuehne, H., Jhuang, H., Garrote, E., Poggio, T., Serre, T.: HMDB: a large video database for human motion recognition. In: ICCV (2011)Google Scholar
  29. 29.
    Larsson, G., Maire, M., Shakhnarovich, G.: Colorization as a proxy task for visual understanding. In: CVPR (2017)Google Scholar
  30. 30.
    Lee, H.Y., Huang, J.B., Singh, M., Yang, M.H.: Unsupervised representation learning by sorting sequences. In: ICCV (2017)Google Scholar
  31. 31.
    Li, Y., Vasconcelos, N.: Repair: removing representation bias by dataset resampling. In: CVPR (2019)Google Scholar
  32. 32.
    Li, Y., Li, Y., Vasconcelos, N.: Resound: towards action recognition without representation bias. In: ECCV (2018)Google Scholar
  33. 33.
    Luo, Z., Zou, Y., Hoffman, J., Fei-Fei, L.F.: Label efficient learning of transferable representations across domains and tasks. In: NeurIPS (2017)Google Scholar
  34. 34.
    Misra, I., Zitnick, C.L., Hebert, M.: Shuffle and learn: unsupervised learning using temporal order verification. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9905, pp. 527–544. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-46448-0_32CrossRefGoogle Scholar
  35. 35.
    Munro, J., Damen, D.: Multi-modal domain adaptation for fine-grained action recognition. In: CVPR (2020)Google Scholar
  36. 36.
    Noroozi, M., Favaro, P.: Unsupervised learning of visual representations by solving jigsaw puzzles. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9910, pp. 69–84. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-46466-4_5CrossRefGoogle Scholar
  37. 37.
    Pan, B., Cao, Z., Adeli, E., Niebles, J.C.: Adversarial cross-domain action recognition with co-attention. In: AAAI (2020)Google Scholar
  38. 38.
    Pathak, D., Krahenbuhl, P., Donahue, J., Darrell, T., Efros, A.A.: Context encoders: Feature learning by inpainting. In: CVPR (2016)Google Scholar
  39. 39.
    Ren, Z., Jae Lee, Y.: Cross-domain self-supervised multi-task feature learning using synthetic imagery. In: CVPR (2018)Google Scholar
  40. 40.
    Saito, K., Watanabe, K., Ushiku, Y., Harada, T.: Maximum classifier discrepancy for unsupervised domain adaptation. In: CVPR (2018)Google Scholar
  41. 41.
    Sikka, K., Sharma, G.: Discriminatively trained latent ordinal model for video classification. TPAMI 40(8), 1829–1844 (2017)CrossRefGoogle Scholar
  42. 42.
    Simonyan, K., Zisserman, A.: Two-stream convolutional networks for action recognition in videos. In: NeurIPS (2014)Google Scholar
  43. 43.
    Soomro, K., Zamir, A.R., Shah, M.: UCF101: A dataset of 101 human actions classes from videos in the wild. arXiv preprint arXiv:1212.0402 (2012)
  44. 44.
    Tran, D., Bourdev, L., Fergus, R., Torresani, L., Paluri, M.: Learning spatiotemporal features with 3d convolutional networks. In: ICCV (2015)Google Scholar
  45. 45.
    Tran, D., Wang, H., Torresani, L., Ray, J., LeCun, Y., Paluri, M.: A closer look at spatiotemporal convolutions for action recognition. In: CVPR (2017)Google Scholar
  46. 46.
    Tsai, Y.H., Sohn, K., Schulter, S., Chandraker, M.: Domain adaptation for structured output via discriminative representations. In: ICCV (2019)Google Scholar
  47. 47.
    Tzeng, E., Hoffman, J., Darrell, T., Saenko, K.: Simultaneous deep transfer across domains and tasks. In: ICCV (2015)Google Scholar
  48. 48.
    Tzeng, E., Hoffman, J., Saenko, K., Darrell, T.: Adversarial discriminative domain adaptation. In: CVPR (2017)Google Scholar
  49. 49.
    Vu, T.H., Jain, H., Bucher, M., Cord, M., Pérez, P.: Dada: Depth-aware domain adaptation in semantic segmentation. In: ICCV (2019)Google Scholar
  50. 50.
    Wang, J., Wang, W., Huang, Y., Wang, L., Tan, T.: M3: multimodal memory modelling for video captioning. In: CVPR (2018)Google Scholar
  51. 51.
    Wang, X., Girshick, R., Gupta, A., He, K.: Non-local neural networks. In: CVPR (2018)Google Scholar
  52. 52.
    Wang, Y., Hoai, M.: Pulling actions out of context: explicit separation for effective combination. In: CVPR (2018)Google Scholar
  53. 53.
    Xie, S., Sun, C., Huang, J., Tu, Z., Murphy, K.: Rethinking spatiotemporal feature learning for video understanding. In: ECCV (2018)Google Scholar
  54. 54.
    Xu, D., Xiao, J., Zhao, Z., Shao, J., Xie, D., Zhuang, Y.: Self-supervised spatiotemporal learning via video clip order prediction. In: CVPR (2019)Google Scholar
  55. 55.
    Xu, D., et al.: Video question answering via gradually refined attention over appearance and motion. In: ACM MM (2017)Google Scholar
  56. 56.
    Zhang, J., Li, W., Ogunbona, P.: Joint geometrical and statistical alignment for visual domain adaptation. In: CVPR (2017)Google Scholar
  57. 57.
    Zhang, Q., Zhang, J., Liu, W., Tao, D.: Category anchor-guided unsupervised domain adaptation for semantic segmentation. In: NeurIPS (2019)Google Scholar
  58. 58.
    Zhou, B., Andonian, A., Oliva, A., Torralba, A.: Temporal relational reasoning in videos. In: ECCV (2018)Google Scholar
  59. 59.
    Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., Torralba, A.: Learning deep features for discriminative localization. In: CVPR (2016)Google Scholar
  60. 60.
    Zhu, X., Pang, J., Yang, C., Shi, J., Lin, D.: Adapting object detectors via selective cross-domain alignment. In: CVPR (2019)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Virginia TechBlacksburgUSA
  2. 2.NEC Labs AmericaSan JoseUSA

Personalised recommendations