Abstract
In the recent years, special emphasis has been placed on visual-based gait recognition due to its unique characteristics such as not requiring a special user action, or its long-distance recognizability. In general, there exist two methods - model-based and appearance-based methods - both of which come with their own advantages and disadvantages. In an effort to harness the best of both worlds we create a compact 3D human model-based gait representation out of 2D images with the help of the DensePose algorithm. We design a simple CNN and train several instances to show that the obtained gait representation can in fact be used to improve gait recognition accuracy. Experimental results are based on the publicly available CASIA-B dataset.
The research reported in this paper has been partly supported by the LIT Secure and Correct Systems Lab funded by the State of Upper Austria and by the Austrian Ministry for Transport, Innovation and Technology, the Federal Ministry of Science, Research and Economy, and the Province of Upper Austria in the frame of the COMET center SCCH.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Chen, X., Weng, J., Lu, W., Xu, J.: Multi-gait recognition based on attribute discovery. IEEE Trans. Pattern Anal. Mach. Intell. 40(7), 1697–1710 (2018)
Güler, R.A., Neverova, N., Kokkinos, I.: DensePose: dense human pose estimation in the wild. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 7297–7306 (2018)
Han, J., Bhanu, B.: Individual recognition using gait energy image. IEEE Trans. Pattern Anal. Mach. Intell. 28(2), 316–322 (2006)
Huynh-The, T., Hua, C.H., Tu, N.A., Kim, D.S.: Learning 3D spatiotemporal gait feature by convolutional network for person identification. Neurocomputing 397, 192–202 (2020)
Krzeszowski, T., Michalczuk, A., Kwolek, B., Switonski, A., Josinski, H.: Gait recognition based on marker-less 3D motion capture. In: 2013 10th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2013, pp. 232–237 (2013)
Saito, S., Huang, Z., Natsume, R., Morishima, S., Li, H., Kanazawa, A.: PIFu: pixel-aligned implicit function for high-resolution clothed human digitization. In: Proceedings of the IEEE International Conference on Computer Vision, October 2019, pp. 2304–2314 (2019)
Wan, C., Wang, L., Phoha, V.V.: A survey on gait recognition. ACM Comput. Surv. 51(5), 1–35 (2018)
Xu, W., Yu, Z., Wang, Z., Guo, B., Han, Q.: AcousticID. Proc. ACM Interact. Mobile Wearable Ubiquit. Technol. 3(3), 1–25 (2019)
Yu, S., Tan, D., Tan, T.: A framework for evaluating the effect of view angle, clothing and carrying condition on gait recognition. In: Proceedings - International Conference on Pattern Recognition, vol. 4, pp. 441–444 (2006)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Schwarz, P., Scharinger, J., Hofer, P. (2022). Gait Recognition with DensePose Energy Images. In: Rozinaj, G., Vargic, R. (eds) Systems, Signals and Image Processing. IWSSIP 2021. Communications in Computer and Information Science, vol 1527. Springer, Cham. https://doi.org/10.1007/978-3-030-96878-6_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-96878-6_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-96877-9
Online ISBN: 978-3-030-96878-6
eBook Packages: Computer ScienceComputer Science (R0)