Abstract
Human pose estimation has achieved significant improvement. However, most existing methods mainly consider how to improve the model performance using complex architecture or computationally expensive model, ignoring the deployment costs in practice, especially in human-robot interaction. In this paper, we investigate a highly efficient pose estimation model with comparable accuracy. We propose an adaptive convolution, which can adaptively generate one or more feature maps with desired channels. Since redundant information in the feature map is an important characteristic, to preserve the redundant information while taking only a few numbers of FLOPs and parameters, we propose a light-weight block based on adaptive convolution, which is performed with two parallel convolution operations. And then, to further reduce the FLOPs, we propose heterogeneous filters based light-weight block, which contains two different kinds of filters in each layer. Finally, three light-weight units are designed to stack light-weight block, and a simple light-weight pose estimation network (SLPE) can be easily established. Extensive evaluations demonstrate the advantages of SLPE over state-of-the-art methods in terms of model cost-effectiveness on the standard benchmark datasets, MPII and COCO dataset.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
If one parallel step is converted to multiple sequential steps, it means increasing the latency. Because all computations have to be done sequentially across layers, the latter layer needs to be executed after the previous layer is executed.
References
Andriluka, M., Pishchulin, L., Gehler, P., Schiele, B.: 2d human pose estimation: new benchmark and state of the art analysis. In: CVPR, pp. 3686–3693 (2014)
Bulat, A., Tzimiropoulos, G.: Binarized convolutional landmark localizers for human pose estimation and face alignment with limited resources. In: ICCV, pp. 3706–3714 (2017)
Cao, Z., Simon, T., Wei, S.-E., Sheikh, Y.: Realtime multi-person 2d pose estimation using part affinity fields. In: CVPR, pp. 7291–7299 (2017)
Chen, Y., Wang, Z., Peng, Y., Zhang, Z., Yu, G., Sun, J.: Cascaded pyramid network for multi-person pose estimation. In: CVPR, pp. 7103–7112 (2018)
Chen, Y., et al.: Drop an octave: reducing spatial redundancy in convolutional neural networks with octave convolution. In: ICCV, pp. 3435–3444 (2019)
Chéron, G., Laptev, I., Schmid, C.: P-CNN: pose-based CNN features for action recognition. In: ICCV, pp. 3218–3226 (2015)
Chollet, F.: Xception: deep learning with depthwise separable convolutions. In: CVPR, pp. 1251–1258 (2017)
Fang, H.-S., Xie, S., Tai, Y.-W., Lu, C.: RMPE: regional multi-person pose estimation. In: ICCV, pp. 2334–2343 (2017)
Han, S., Mao, H., Dally, W.J.: Deep compression: compressing deep neural networks with pruning, trained quantization and Huffman coding. In: ICLR, pp. 1–14 (2016)
He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask R-CNN. In: ICCV, pp. 2961–2969 (2017)
He, Y., Kang, G., Dong, X., Fu, Y., Yang, Y.: Soft filter pruning for accelerating deep convolutional neural networks. In: IJCAI, pp. 2234–2240 (2018)
Howard, A.G., et al.: MobileNets: efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861 (2017)
Huang, S., Gong, M., Tao, D.: A coarse-fine network for keypoint localization. In: ICCV, pp. 3028–3037 (2017)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: ICLR, pp. 1–15 (2015)
Kocabas, M., Karagoz, S., Akbas, E.: MultiPoseNet: fast multi-person pose estimation using pose residual network. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11215, pp. 437–453. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01252-6_26
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: NIPS, pp. 1097–1105 (2012)
Lin, T.-Y., et al.: Microsoft COCO: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8693, pp. 740–755. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10602-1_48
Ma, N., Zhang, X., Zheng, H.-T., Sun, J.: ShuffleNet V2: practical guidelines for efficient CNN architecture design. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) Computer Vision – ECCV 2018. LNCS, vol. 11218, pp. 122–138. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01264-9_8
Newell, A., Yang, K., Deng, J.: Stacked hourglass networks for human pose estimation. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9912, pp. 483–499. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46484-8_29
Sun, K., Xiao, B., Liu, D., Wang, J.: Deep high-resolution representation learning for human pose estimation. In: CVPR, pp. 5693–5703 (2019)
Tang, W., Yu, P., Wu, Y.: Deeply learned compositional models for human pose estimation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11207, pp. 197–214. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01219-9_12
Wan, B., Zhou, D., Liu, Y., Li, R., He, X.: Pose-aware multi-level feature network for human object interaction detection. In: ICCV, pp. 9469–9478 (2019)
Wang, C., Wang, Y., Yuille, A.L.: An approach to pose-based action recognition. In: CVPR, pp. 915–922 (2013)
Xiao, B., Wu, H., Wei, Y.: Simple baselines for human pose estimation and tracking. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11210, pp. 472–487. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01231-1_29
Yan, S., Xiong, Y., Lin, D.: Spatial temporal graph convolutional networks for skeleton-based action recognition. In: AAAI, pp. 7444–7452 (2018)
Yang, W., Li, S., Ouyang, W., Li, H., Wang, X.: Learning feature pyramids for human pose estimation. In: ICCV, pp. 1281–1290 (2017)
Zhang, X., Zhou, X., Lin, M., Sun, J.: ShuffleNet: an extremely efficient convolutional neural network for mobile devices. In: CVPR, pp. 6848–6856 (2018)
Howard, A., et al.: Searching for mobilenetv3. In: ICCV, pp. 1314–1324 (2019)
Andrew, H., Andrey, Z., Liang-Chieh, C., Mark, S., Menglong, Z.: Inverted residuals and linear bottlenecks: mobile networks for classification, detection and segmentation. In: CVPR, pp. 122–138 (2018)
Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., Liu, Z.: Dynamic convolution: attention over convolution kernels. In: CVPR, pp. 11030–11039 (2020)
Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., Liu, Z.: Dynamic ReLU. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12364, pp. 351–367. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58529-7_21
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Sun, B., Zhao, M. (2021). Simple Light-Weight Network for Human Pose Estimation. In: Pham, D.N., Theeramunkong, T., Governatori, G., Liu, F. (eds) PRICAI 2021: Trends in Artificial Intelligence. PRICAI 2021. Lecture Notes in Computer Science(), vol 13033. Springer, Cham. https://doi.org/10.1007/978-3-030-89370-5_21
Download citation
DOI: https://doi.org/10.1007/978-3-030-89370-5_21
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-89369-9
Online ISBN: 978-3-030-89370-5
eBook Packages: Computer ScienceComputer Science (R0)