Abstract
Most existing action recognition models are large convolutional neural networks that work only with raw RGB frames as input. However, practical applications require lightweight models that directly process compressed videos. In this work, for the first time, such a model is developed, which is lightweight enough to run in real-time on embedded AI devices without sacrifices in recognition accuracy. A new Aligned Temporal Trilinear Pooling (ATTP) module is formulated to fuse three modalities in a compressed video. To remedy the weaker motion vectors (compared to optical flow computed from raw RGB streams) for representing dynamic content, we introduce a temporal fusion method to explicitly induce the temporal context, as well as knowledge distillation from a model trained with optical flows via feature alignment. Compared to existing compressed video action recognition models, it is much more compact and faster thanks to adopting a lightweight CNN backbone.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Carreira, J., Zisserman, A.: Quo vadis, action recognition? a new model and the kinetics dataset. In: CVPR, pp. 6299–6308 (2017)
Chollet, F.: Xception: deep learning with depthwise separable convolutions. In: CVPR, pp. 1251–1258 (2017)
Ding, M., et al.: Learning depth-guided convolutions for monocular 3D object detection. In: CVPR, pp. 4306–4315 (2020)
Ding, M., Wang, Z., Sun, J., Shi, J., Luo, P.: Camnet: coarse-to-fine retrieval for camera re-localization. In: ICCV, pp. 2871–2880 (2019)
Ding, M., Wang, Z., Zhou, B., Shi, J., Lu, Z., Luo, P.: Every frame counts: joint learning of video segmentation and optical flow. In: AAAI, pp. 10713–10720 (2020)
Ding, M., Zhao, A., Lu, Z., Xiang, T., Wen, J.R.: Face-focused cross-stream network for deception detection in videos. In: CVPR, pp. 7802–7811 (2019)
Feichtenhofer, C., Fan, H., Malik, J., He, K.: Slowfast networks for video recognition. In: ICCV, pp. 6202–6211 (2019)
Forecast, C.V.: Cisco visual networking index: Forecast and trends, 2017–2022. Cisco Public Information, White paper (2019)
Gao, Y., Beijbom, O., Zhang, N., Darrell, T.: Compact bilinear pooling. In: CVPR, pp. 317–326 (2016)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR, pp. 770–778 (2016)
Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 (2015)
Howard, A., et al.: Searching for mobilenetv3. In: ICCV, pp. 1314–1324 (2019)
Huo, Y., Xu, X., Lu, Y., Niu, Y., Lu, Z., Wen, J.R.: Mobile video action recognition. arXiv preprint arXiv:1908.10155 (2019)
Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., Keutzer, K.: Squeezenet: alexnet-level accuracy with 50x fewer parameters and \(<\) 0.5 mb model size. arXiv preprint arXiv:1602.07360 (2016)
Ji, J., Buch, S., Soto, A., Carlos Niebles, J.: End-to-end joint semantic segmentation of actors and actions in video. In: ECCV, pp. 702–717 (2018)
Kantorov, V., Laptev, I.: Efficient feature extraction, encoding and classification for action recognition. In: CVPR, pp. 2593–2600 (2014)
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., Fei-Fei, L.: Large-scale video classification with convolutional neural networks. In: CVPR, pp. 1725–1732 (2014)
Kay, W., et al.: The kinetics human action video dataset. arXiv preprint arXiv:1705.06950 (2017)
Kim, J.H., On, K.W., Lim, W., Kim, J., Ha, J.W., Zhang, B.T.: Hadamard product for low-rank bilinear pooling. In: ICLR (2016)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: NeurIPS, pp. 1097–1105 (2012)
Kuehne, H., Jhuang, H., Garrote, E., Poggio, T., Serre, T.: Hmdb: a large video database for human motion recognition. In: ICCV, pp. 2556–2563. IEEE (2011)
Lin, J., Gan, C., Han, S.: Tsm: temporal shift module for efficient video understanding. In: ICCV, pp. 7083–7093 (2019)
Lin, T., Zhao, X., Su, H., Wang, C., Yang, M.: Bsn: boundary sensitive network for temporal action proposal generation. In: ECCV, pp. 3–19 (2018)
Lin, T.Y., RoyChowdhury, A., Maji, S.: Bilinear cnn models for fine-grained visual recognition. In: ICCV, pp. 1449–1457 (2015)
Ma, N., Zhang, X., Zheng, H.T., Sun, J.: Shufflenet v2: practical guidelines for efficient cnn architecture design. In: ECCV, pp. 116–131 (2018)
Shou, Z., et al.: Dmc-net: Generating discriminative motion cues for fast compressed video action recognition. In: CVPR, pp. 1268–1277 (2019)
Sikora, T.: The mpeg-4 video standard verification model. TCSV 7(1), 19–31 (1997)
Simonyan, K., Zisserman, A.: Two-stream convolutional networks for action recognition in videos. In: NeurIPS, pp. 568–576 (2014)
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)
Sullivan, G.J., Ohm, J.R., Han, W.J., Wiegand, T.: Overview of the high efficiency video coding (HEVC) standard. TCSV 22(12), 1649–1668 (2012)
Tan, M., Le, Q.V.: Efficientnet: rethinking model scaling for convolutional neural networks. In: ICML, pp. 6105–6114 (2019)
Tran, D., Bourdev, L., Fergus, R., Torresani, L., Paluri, M.: Learning spatiotemporal features with 3D convolutional networks. In: ICCV, pp. 4489–4497 (2015)
Tran, D., Wang, H., Torresani, L., Ray, J., LeCun, Y., Paluri, M.: A closer look at spatiotemporal convolutions for action recognition. In: CVPR, pp. 6450–6459 (2018)
Wang, L., et al.: Temporal segment networks: towards good practices for deep action recognition. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9912, pp. 20–36. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46484-8_2
Wang, X., Girshick, R., Gupta, A., He, K.: Non-local neural networks. In: CVPR, pp. 7794–7803 (2018)
Wiegand, T., Sullivan, G.J., Bjontegaard, G., Luthra, A.: Overview of the H. 264/ AVC video coding standard. TCSV 13(7), 560–576 (2003)
Wu, C.Y., Zaheer, M., Hu, H., Manmatha, R., Smola, A.J., Krähenbühl, P.: Compressed video action recognition. In: CVPR, pp. 6026–6035 (2018)
Yu, Z., Yu, J., Fan, J., Tao, D.: Multi-modal factorized bilinear pooling with co-attention learning for visual question answering. In: ICCV, pp. 1821–1830 (2017)
Zhang, B., Wang, L., Wang, Z., Qiao, Y., Wang, H.: Real-time action recognition with enhanced motion vector CNNs. In: CVPR, pp. 2718–2726 (2016)
Zhang, B., Wang, L., Wang, Z., Qiao, Y., Wang, H.: Real-time action recognition with deeply transferred motion vector CNNs. TIP 27(5), 2326–2339 (2018)
Zolfaghari, M., Singh, K., Brox, T.: Eco: efficient convolutional network for online video understanding. In: ECCV, pp. 695–712 (2018)
Acknowledgements
This work was supported by Beijing Outstanding Young Scientist Program (BJJWZYJH012019100020098), National Natural Science Foundation of China (61976220 and 61832017), and the Outstanding Innovative Talents Cultivation Funded Programs 2018 of Renmin University of China.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Huo, Y. et al. (2020). Lightweight Action Recognition in Compressed Videos. In: Bartoli, A., Fusiello, A. (eds) Computer Vision – ECCV 2020 Workshops. ECCV 2020. Lecture Notes in Computer Science(), vol 12536. Springer, Cham. https://doi.org/10.1007/978-3-030-66096-3_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-66096-3_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-66095-6
Online ISBN: 978-3-030-66096-3
eBook Packages: Computer ScienceComputer Science (R0)