Abstract
Still image-based action recognition is a challenging area in which recognition is performed based on only a single input image. Utilizing auxiliary information such as pose, object, or background is one of the common techniques in this field. However, the simultaneous use of several auxiliary components and their optimal combinations is less studied. In this study, two cues of body joints and objects have been employed simultaneously, and an attention module is proposed to combine the features of these two components. The attention module consists of two self-attentions and a cross-attention, which are designed to account for the interaction between the objects, between the joints, and between the joints and objects, respectively. In addition, the Multi-scale Atrous Spatial Pyramid Pooling (MASPP) module is proposed to reduce the number of parameters of the proposed method and at the same time, combine the features obtained from different levels of the backbone. The Joint Object Pooling (JOPool) module is proposed to extract local features from joints and objects regions. ResNets are used as the backbone, and the stride of the last two layers is changed. Experimental results on different datasets show that the combination of several auxiliary components can be effective in increasing the mean Average Precision (mAP) of recognition. The proposed method is evaluated on three important datasets: Stanford-40, PASCAL VOC 2012, and BU101PLUS resulting in 94.84%, 93.20%, and 91.25% mAPs, respectively. The obtained mAPs are higher than the best preceding proposed methods.
Similar content being viewed by others
Data availability
The data that support the findings of this study are available in the reference numbers [2, 13, 49]. These data were derived from the following resources available in the public domain:
• Stanford-40 Actions: http://vision.stanford.edu/Datasets/40actions.html
• PASCAL VOC 2012: http://host.robots.ox.ac.uk/pascal/VOC/voc2012/
• BU101PLUS: https://github.com/seyedsajadashrafi/bu101plus-action-recognition-dataset
References
Akti S, Ofli F, Imran M, Ekenel HK (2021) “Fight Detection from Still Images in the Wild,” Proc. - 2022 IEEE/CVF Winter Conf. Appl. Comput. Vis. Work. WACVW 2022, pp. 550–559, https://doi.org/10.48550/arxiv.2111.08370
Ashrafi SS, Shokouhi SB, Ayatollahi A (Jul. 2021) Action recognition in still images using a multi-attention guided network with weakly supervised saliency detection. Multimed Tools Appl 2021:1–27. https://doi.org/10.1007/S11042-021-11215-1
Beddiar DR, Nini B, Sabokrou M, Hadid A (2020) Vision-based human activity recognition: a survey. Multimed Tools Appl 79:1–47. https://doi.org/10.1007/s11042-020-09004-3
Cao Y, Liu C, Huang Z, Sheng Y, Ju Y (Jun. 2021) Skeleton-based action recognition with temporal action graph and temporal adaptive graph convolution structure. Multimed Tools Appl 2021:1–24. https://doi.org/10.1007/S11042-021-11136-Z
Chakraborty S, Mondal R, Singh PK, Sarkar R, Bhattacharjee D (2021) Transfer learning with fine tuning for human action recognition from still images. Multimed Tools Appl 2021 8013 80(13):20547–20578. https://doi.org/10.1007/S11042-021-10753-Y
Chapariniya M, Ashrafi SS, Shokouhi SB (2020) “Knowledge Distillation Framework for Action Recognition in Still Images”, 2020 10h Int. Conf Comput Knowl Eng ICCKE 2020, pp. 274–277, https://doi.org/10.1109/ICCKE50421.2020.9303716
Chen LC, Papandreou G, Kokkinos I, Murphy K, Yuille AL (2016) “DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 40, no. 4, pp. 834–848, Accessed: Aug. 12, 2021. [Online]. Available: https://arxiv.org/abs/1606.00915v2
Chollet F (2016) “Xception: Deep Learning with Depthwise Separable Convolutions,” Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, vol. 2017-January, pp. 1800–1807, Accessed: Aug. 11, 2021. [Online]. Available: https://arxiv.org/abs/1610.02357v3
Chu J, Guo Z, Leng L (Mar. 2018) Object detection based on multi-layer convolution feature fusion and online hard example mining. IEEE Access 6:19959–19967. https://doi.org/10.1109/ACCESS.2018.2815149
Dehkordi HA, Nezhad AS, Ashrafi SS, Shokouhi SB (2021) “Still Image Action Recognition Using Ensemble Learning,” 2021 7th Int. Conf Web Res ICWR 2021, pp. 125–129, https://doi.org/10.1109/ICWR51868.2021.9443021
Dehkordi HA, Nezhad AS, Kashiani H, Shokouhi SB, Ayatollahi A (2022) “Multi-expert human action recognition with hierarchical super-class learning”, Knowledge-Based Syst., p. 109091, https://doi.org/10.1016/J.KNOSYS.2022.109091
Dosovitskiy A et al. (2020) “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale”, Accessed: Aug. 12, 2021. [Online]. Available: https://arxiv.org/abs/2010.11929v2
Everingham M, Van Gool L, Williams CKI, Winn J, Zisserman A (Jun. 2010) The pascal visual object classes (VOC) challenge. Int J Comput Vis 88(2):303–338. https://doi.org/10.1007/s11263-009-0275-4
Gkioxari G, Girshick R, Malik J (2015) “Contextual action recognition with R∗CNN,” Proc. IEEE Int. Conf. Comput. Vis., vol. 2015 Inter, pp 1080–1088 https://doi.org/10.1109/ICCV.2015.129
Guo G, Lai A (2014) A survey on still image based human action recognition. Pattern Recogn 47(10):3343–3361. https://doi.org/10.1016/j.patcog.2014.04.018
He K, Zhang X, Ren S, Sun J (2016) “Deep residual learning for image recognition,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2016-December, pp. 770–778, https://doi.org/10.1109/CVPR.2016.90
He K, Gkioxari G, Dollár P, Girshick R (Feb. 2020) Mask R-CNN. IEEE Trans Pattern Anal Mach Intell 42(2):386–397. https://doi.org/10.1109/TPAMI.2018.2844175
Herath S, Harandi M, Porikli F (2017) Going deeper into action recognition: a survey. Image Vis Comput 60:4–21. https://doi.org/10.1016/j.imavis.2017.01.010
Hinton G, Vinyals O, Dean J (2015) “Distilling the Knowledge in a Neural Network”, Accessed: Aug. 11, 2021. [Online]. Available: https://arxiv.org/abs/1503.02531v1.
Hu T, Zhu X, Guo W, Wang S, Zhu J (Feb. 2018) Human action recognition based on scene semantics. Multimed Tools Appl 2018 7820 78(20):28515–28536. https://doi.org/10.1007/S11042-017-5496-X
Kim S, Yun K, Park J, Choi JY (2019) “Skeleton-based Action Recognition of People Handling Objects”, Proc. - 2019 IEEE Winter Conf. Appl. Comput. Vision, WACV 2019, pp. 61–70, Accessed: Aug. 13, 2021. [Online]. Available: https://arxiv.org/abs/1901.06882v1
Kipf TN, Welling M(2016) “Semi-Supervised Classification with Graph Convolutional Networks,” 5th Int. Conf. Learn. Represent. ICLR 2017 - Conf. Track Proc., Accessed: Aug. 13, 2021. [Online]. Available: https://arxiv.org/abs/1609.02907v4
Li LJ, Fei-Fei L (2007) “What, where and who? Classifying events by scene and object recognition”, https://doi.org/10.1109/ICCV.2007.4408872
Li Y, Li K, Wang X (Aug. 2020) Recognizing actions in images by fusing multiple body structure cues. Pattern Recogn 104:107341. https://doi.org/10.1016/j.patcog.2020.107341
Liao X, Li K, Zhu X, Liu KJR (Aug. 2020) Robust detection of image operator chain with two-stream convolutional neural network. IEEE J Sel Top Signal Proc 14(5):955–968. https://doi.org/10.1109/JSTSP.2020.3002391
Liu L, Tan RT, You S (2019) “Loss Guided Activation for Action Recognition in Still Images”, in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11365 LNCS, pp. 152–167, https://doi.org/10.1007/978-3-030-20873-8_10
Ludl D, Gulde T, Curio C (2019) “Simple yet efficient real-time pose-based action recognition”, in 2019 IEEE Intelligent Transportation Systems Conference, ITSC 2019, pp. 581–588, https://doi.org/10.1109/ITSC.2019.8917128
Ma W, Liang S (2020) “Human-object relation network for action recognition in still images”, Proc. - IEEE Int. Conf. Multimed. Expo, vol. 2020-July, https://doi.org/10.1109/ICME46284.2020.9102933.
Ma S, Bargal SA, Zhang J, Sigal L, Sclaroff S (Aug. 2017) Do less and achieve more: training CNNs for action recognition utilizing action images from the web. Pattern Recogn 68:334–345. https://doi.org/10.1016/j.patcog.2017.01.027
Maji S, Bourdev L, Malik J “Action Recognition from a Distributed Representation of Pose and Appearance”
McAuley J, Leskovec J (2012) “Image labeling on a network: Using social-network metadata for image classification,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7575 LNCS, no. PART 4, pp. 828–841, https://doi.org/10.1007/978-3-642-33765-9_59.
Mi S, Zhang Y (2021) Pose-guided action recognition in static images using lie-group. Appl Intell 2021:1–9. https://doi.org/10.1007/S10489-021-02760-1
Mohammadi S, Majelan SG, Shokouhi SB (2019) “Ensembles of deep neural networks for action recognition in still images”, 2019 9th Int. Conf. Comput. Knowl. Eng. ICCKE 2019, pp. 315–318, https://doi.org/10.1109/ICCKE48569.2019.8965014
Procesi C (2007) “Lie groups : an approach through invariants and representations,” p. 596
Qi T, Xu Y, Quan Y, Wang Y, Ling H (Dec. 2017) Image-based action recognition using hint-enhanced deep neural networks. Neurocomputing 267:475–488. https://doi.org/10.1016/j.neucom.2017.06.041
Ren S, He K, Girshick R, Sun J (2017) Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39(6):1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031
Ren Z, Zhang Q, Gao X, Hao P, Cheng J (Mar. 2020) Multi-modality learning for human action recognition. Multimed Tools Appl 2020 8011 80(11):16185–16203. https://doi.org/10.1007/S11042-019-08576-Z
Simonyan K, Zisserman A, “Two-Stream Convolutional Networks for Action Recognition in Videos.”
Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z (2015) “Rethinking the Inception Architecture for Computer Vision”, Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-December, pp. 2818–2826, Accessed: Aug. 11, 2021. [Online]. Available: https://arxiv.org/abs/1512.00567v3.
Szegedy C et al. (2015) “Going deeper with convolutions”, in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 07–12-June-2015, pp. 1–9, https://doi.org/10.1109/CVPR.2015.7298594
C. Szegedy, S. Ioffe, V. Vanhoucke, and A. Alemi (2016) “Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning,” 31st AAAI Conf. Artif. Intell. AAAI 2017, pp. 4278–4284, Accessed: Aug. 11, 2021. [Online]. Available: https://arxiv.org/abs/1602.07261v2
Tran D, Wang H, Torresani L, Ray J, LeCun Y, Paluri M (2017) “A Closer Look at Spatiotemporal Convolutions for Action Recognition,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., pp. 6450–6459, Accessed: Aug. 13, 2021. [Online]. Available: https://arxiv.org/abs/1711.11248v3
Wang J, Liang S, “Pose-Enhanced Relation Feature for Action Recognition in Still Images” (2022) pp. 154–165, https://doi.org/10.1007/978-3-030-98358-1_13
Wang X, Qi C (Dec. 2019) Detecting action-relevant regions for action recognition using a three-stage saliency detection technique. Multimed Tools Appl 2019 7911 79(11):7413–7433. https://doi.org/10.1007/S11042-019-08535-8
Wang C, Yang H, Meinel C (2016) “Exploring multimodal video representation for action recognition,” Proc. Int. Jt. Conf. Neural Networks, vol. 2016-October, pp. 1924–1931, https://doi.org/10.1109/IJCNN.2016.7727435
Xin M, Wang S, Cheng J (2019) “Entanglement loss for context-based still image action recognition,” in Proceedings - IEEE International Conference on Multimedia and Expo, vol. 2019-July, pp. 1042–1047, https://doi.org/10.1109/ICME.2019.00183
Xu Y, Hou Z, Liang J, Chen C, Jia L, Song Y (May 2019) Action recognition using weighted fusion of depth images and skeleton’s key frames. Multimed Tools Appl 2019 7817 78(17):25063–25078. https://doi.org/10.1007/S11042-019-7593-5
Yan S, Smith JS, Lu W, Zhang B (Dec. 2018) Multibranch attention networks for action recognition in still images. IEEE Trans Cogn Dev Syst 10(4):1116–1125. https://doi.org/10.1109/TCDS.2017.2783944
Yao B, Jiang X, Khosla A, Lin AL, Guibas L, Fei-Fei L (2011) “Human action recognition by learning bases of action attributes and parts,” in Proceedings of the IEEE International Conference on Computer Vision, pp. 1331–1338, https://doi.org/10.1109/ICCV.2011.6126386
Zhang Y, Chu J, Leng L, Miao J (2020) Mask-Refined R-CNN: A Network for Refining Object Details in Instance Segmentation. Sensors (Basel) 20(4). https://doi.org/10.3390/S20041010
Zhao Z, Ma H, You S (2017) “Single Image Action Recognition Using Semantic Body Part Actions,” in Proceedings of the IEEE International Conference on Computer Vision, vol. 2017-October, pp. 3411–3419, https://doi.org/10.1109/ICCV.2017.367
Zheng Y, Zheng X, Lu X, Wu S (Nov. 2020) Spatial attention based visual semantic learning for action recognition in still images. Neurocomputing 413:383–396. https://doi.org/10.1016/J.NEUCOM.2020.07.016
Zhu Y et al. (2020) “A Comprehensive Study of Deep Video Action Recognition”, Accessed: Aug. 12, 2021. [Online]. Available: https://arxiv.org/abs/2012.06567v1.
Zoph B, Vasudevan V, Shlens J, Le QV (2017) “Learning Transferable Architectures for Scalable Image Recognition”, Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., pp. 8697–8710, Accessed: Aug. 11, 2021. [Online]. Available: https://arxiv.org/abs/1707.07012v4.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that there is no conflict of interest.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Ashrafi, S.S., Shokouhi, S.B. & Ayatollahi, A. Still image action recognition based on interactions between joints and objects. Multimed Tools Appl 82, 25945–25971 (2023). https://doi.org/10.1007/s11042-023-14350-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-14350-z