Abstract
Action recognition in still images is a popular research topic in the field of computer vision, but it is to remain challenging due to the lack of motion information. Contextual information is a significant factor in the task of recognizing image action, which is inseparable from a predefined action class. And the existing research strategy does not ensure adequate use of contextual information. To address this issue, we propose a Contextual Enhancement Module (CEM) that combines the self-attention mechanism and the contextual attention mechanism. Specifically, the context enhancement module uses self-attention to learn pixel-level contextual information, after which separates the image into parts and uses contextual attention to learn region-level contextual information. In this way, the model can emphasize the significance of various pixels and regions in the image and significantly improve feature representation. We performed a lot of experiments on the PASCAL VOC 2012 Action dataset and the Stanford 40 Actions dataset. The results demonstrate that our method performs effectively, with the state-of-the-arts outcomes being obtained on both datasets.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Zhu, Y., et al.: A comprehensive study of deep video action recognition. arXiv preprint arXiv:2012.06567 (2020)
Girish, D., Singh, V., Ralescu, A.: Understanding action recognition in still images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pp. 370–371 (2020)
Dosovitskiy, A., et al.: An image is worth 16 x 16 words: transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 (2020)
Vaswani, A., et al.: Attention is all you need. Adv. Neural Inf. Process. Syst. 30 (2017)
Behera, A., Wharton, Z., Hewage, P.R., Bera, A.: Context-aware attentional pooling (cap) for fine-grained visual classification. Proc.AAAI Conf. Artif. Intell. 35(2), 929–937 (2021)
Wang, Y, et al.: Unsupervised discovery of action classes. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2006), vol. 2. IEEE (2006)
Gkioxari, G., Girshick, R., Malik, J.: Contextual action recognition with r* CNN. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1080–1088 (2015)
Zhao, Z., Ma, H., Chen, X.: Semantic parts based top-down pyramid for action recognition. Patt. Recogn. Lett. 84, 134–141 (2016)
Zhao, Z., Ma, H., You, S.: Single image action recognition using semantic body part actions. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 3391–3399 (2017)
Yan, S., Smith, J.S., Lu, W., Zhang, B.: Multibranch attention networks for action recognition in still images. IEEE Trans. Cognitive Dev. Syst. 10(4), 1116–1125 (2017)
Zhu, H., Hu, J.F., Zheng, W.S.: Learning hierarchical context for action recognition in still images. In: Advances in Multimedia Information Processing–PCM 2018: 19th Pacific-Rim Conference on Multimedia, Hefei, China, 21–22 September, 2018, Proceedings, Part III 19, pp. 67–77 (2018)
Zheng, Y., Zheng, X., Lu, X., Wu, S.: Spatial attention based visual semantic learning for action recognition in still images. Neurocomputing 413, 383–396 (2020)
Ma, W., Liang, S.: Human-object relation network for action recognition in still images. In: 2020 IEEE International Conference on Multimedia and Expo (ICME), pp. 1–6 (2020)
Wang, J., Liang, S.: Pose-enhanced relation feature for action recognition in still images. In: Þór Jónsson, B., et al. MultiMedia Modeling. MMM 2022. Lecture Notes in Computer Science, vol. 13141. Springer, Cham (2022). https://doi.org/10.1007/978-3-030-98358-1_13
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Zhang, H., Goodfellow, I., Metaxas, D., Odena, A.: Self-attention generative adversarial networks. In: International Conference on Machine Learning, pp. 7354–7363 (2019)
Yu, C., Zhao, X., Zheng, Q., Zhang, P., You, X.: Hierarchical bilinear pooling for fine-grained visual recognition. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 574–589 (2018)
Everingham, M., Van Gool, L., Williams, C.K., Winn, J., Zisserman, A.: The pascal Visual Object Classes (voc) challenge. Int. J. Comput. Vision 88, 303–338 (2010)
Yao, B., Jiang, X., Khosla, A., Lin, A.L., Guibas, L., Fei-Fei, L.: Human action recognition by learning bases of action attributes and parts. In: 2011 International Conference on Computer Vision, pp. 1331–1338 (2011)
Ruder, S.: An overview of gradient descent optimization algorithms. arXiv preprint arXiv:1609.04747(2016)
Mi, S., Zhang, Y.: Pose-guided action recognition in static images using lie-group. Appl. Intell. 1–9(2022)
Wu, W., Yu, J.: An improved deep relation network for action recognition in still images. In: ICASSP 2021–2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 2450–2454 (2021)
Li, Y., Li, K., Wang, X.: Recognizing actions in images by fusing multiple body structure cues. Patt. Recogn. 104, 107341 (2020)
Acknowledgement
This work is supported by the Inner Mongolia Science and Technology Project (No. 2021GG0166).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
He, J., Wu, W., Li, Y. (2023). Context Enhancement Methodology for Action Recognition in Still Images. In: Iliadis, L., Papaleonidas, A., Angelov, P., Jayne, C. (eds) Artificial Neural Networks and Machine Learning – ICANN 2023. ICANN 2023. Lecture Notes in Computer Science, vol 14254. Springer, Cham. https://doi.org/10.1007/978-3-031-44207-0_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-44207-0_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44206-3
Online ISBN: 978-3-031-44207-0
eBook Packages: Computer ScienceComputer Science (R0)