Abstract
Human recognition at distance has ever been an attractive solution for many applications. However, the challenges associated with this task, especially in the wild, are difficult to overcome. Most gait recognition methods do not take into account the shape of the body as a whole, but also a walking style, which helps to recognize a person more accurately. We offer a method for recognizing walking style at a reasonable distance of the camera so that the visual information of a walking person can be captured and analyzed. Method is based on the latest achievements in deep object detection and tracking, as well as deep networks for human action recognition, adapted for walking style recognition. We study six main categories of walking style based on head position, torso position, arm swing, stride length, and walking speed. Additionally, we analyze several clips from each video sequence for accurate recognition. Experiments show that sometimes we cannot explicitly assign a walking style to one category, but two categories such as driver and influencer describe human walking in the wild well. We achieved recognition results 82.4% Top-1 and 94.7% Top-2.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Adolph, D., Wolfgang Tschacher, W., Niemeyer, H., Michalak. J.: Gait patterns and mood in everyday life: A comparison between depressed patients and non‑depressed controls. Cogn. Therapy Res. 45, 1128–1140 (2021)
Shiguematsu, Y.M., Brandao, M., Takanishi, A.: Effects of walking style and symmetry on the performance of localization algorithms for a biped humanoid robot. In: 2019 IEEE/SICE International Symposium on System Integration (SII), pp. 307–312. IEEE, Paris, France (2019)
Xie, Y., Zheng, J., Hou, X., Xi, Y., Tian, F.: Dynamic dual-peak network: A real-time human detection network in crowded scenes. J. Vis. Commun. Image R. 79, 103195.1–103195.10 (2021)
Fu, K., Zhang, T., Zhang, Y., Sun, X.: OSCD: A one-shot conditional object detection framework. Neurocomputing 425, 243–255 (2021)
Yang, Z., Soltanian-Zadeh, S., Farsiu, S.: BiconNet: An edge-preserved connectivity-based approach for salient object detection. Pattern Recognit. 121, 108231.1–108231.11 (2022)
Zheng, Q., Li, Y., Zheng, L., Shen, Q.: Progressively real-time video salient object detection via cascaded fully convolutional networks with motion attention. Neurocomputing 467, 465–475 (2022)
Danelljan, M, Khan, F, Felsberg, M, Weijer, J.: Learning spatially regularized correlation filters for visual tracking. In: Proceedings of IEEE International Conference on Computer Vision Workshop (ICCV), pp. 4310–4318. IEEE, Santiago, Chile (2015)
Li, P.X., Wang, D., Wang, L.J., Lu, H.C.: Deep visual tracking: review and experimental comparison. Pattern Recognit. 76, 323–338 (2018)
Chi, Z., Li, H., Lu, H., Yang, M.: Dual deep network for visual tracking. IEEE Trans. Image Process. 26(4), 2005–2015 (2017)
Gao, J., Zhang, T., Yang, X., Xu, C.: Deep relative tracking. IEEE Trans. Image Process. 26(4), 1845–1858 (2017)
Bochkovskiy, A., Wang, C.-Y., Liao, H.-Y.M.: YOLOv4: Optimal speed and accuracy of object detection. CoRR arXiv preprint, arXiv:2004.10934v1 (2020)
Geiger, A., Lenz, P. Urtasun, R.: Are we ready for autonomous driving? The KITTI vision benchmark suite. In: IEEE 2012 Conference on Computer Vision and Pattern Recognition (CVPR 2012), pp. 3354–3361. IEEE, Providence, Rhode Island (2012)
Milan, A., Leal-Taixé, L., Reid, I., Roth, S., Schindler, K.: MOT16: A benchmark for multi-object tracking. CoRR arXiv preprint, arXiv: 1603.00831 (2016)
Alotaibi, M., Mahmood, A.: Improved gait recognition based on specialized deep convolutional neural network. Comput. Vis. Image Underst. 164, 103–110 (2017)
Tran, D., Bourdev, L., Fergus, R., Torresani, L., Paluri, M.: Learning spatiotemporal features with 3D convolutional networks. In: The 15th IEEE International Conference on Computer Vision (ICCV), pp. 4489–4497. IEEE, Santiago, Chile (2015)
Carreira, J., Zisserman A.: Quo Vadis, action recognition? A new model and the Kinetics dataset. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4724–4733. IEEE, Honolulu, HI, USA (2017)
WalkingStyles dataset v.211812, https://drive.google.com/drive/folders/1KA392PUfNTnDy-vL6kiDSeZ-Wm6elKhi?usp=sharing. Last accessed 26 Dec 2021
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Favorskaya, M.N., Buryachenko, V.V. (2022). Vision-Based Walking Style Recognition in the Wild. In: Czarnowski, I., Howlett, R.J., Jain, L.C. (eds) Intelligent Decision Technologies. Smart Innovation, Systems and Technologies, vol 309. Springer, Singapore. https://doi.org/10.1007/978-981-19-3444-5_19
Download citation
DOI: https://doi.org/10.1007/978-981-19-3444-5_19
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-3443-8
Online ISBN: 978-981-19-3444-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)