Research Article
BibTex RIS Cite

Analysis of an Image Recognition Method on Group Activities

Year 2020, Ejosat Special Issue 2020 (HORA), 68 - 72, 15.08.2020
https://doi.org/10.31590/ejosat.779063

Abstract

Recognizing group activity on still images is a very challenging problem. The difficulty in a distinction between foreground and background on images makes this problem more complicated than the problem of recognizing group activity on video due to the lack of spatial and temporal information. In this study, we examine the analysis of a still image recognition method on the Volleyball video dataset, which is collected for group activity recognition. We feed an additional mean image that is obtained from the previous and/or subsequent frames with the target image in order to analyze the temporal information gain. We aim to acquire temporal information from the mean images and to use it to train our method. As it is understood from the experimental results, our proposed method can get comparable results with the state-of-the-art video-based group activity recognition studies.

References

  • Ibrahim, M. S., Muralidharan, S., Deng, Z., Vahdat, A., & Mori, G. (2016). A hierarchical deep temporal model for group activity recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1971-1980).
  • Shu, T., Todorovic, S., & Zhu, S. C. (2017). CERN: confidence-energy recurrent network for group activity recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 5523-5531).
  • Qi, M., Qin, J., Li, A., Wang, Y., Luo, J., & Van Gool, L. (2018). stagnet: An attentive semantic RNN for group activity recognition. In Proceedings of the European Conference on Computer Vision (ECCV) (pp. 101-117).
  • Lu, L., Di, H., Lu, Y., Zhang, L., & Wang, S. (2018). A two-level attention-based interaction model for multi-person activity recognition. Neurocomputing, 322, 195-205.
  • Lu, L., Lu, Y., Yu, R., Di, H., Zhang, L., & Wang, S. (2019). GAIM: Graph Attention Interaction Model for Collective Activity Recognition. IEEE Transactions on Multimedia, 22(2), 524-539.
  • Zalluhoglu, C., & Ikizler-Cinbis, N. (2019). Region based multi-stream convolutional neural networks for collective activity recognition. Journal of Visual Communication and Image Representation, 60, 170-179.
  • Choi, W., Chao, Y. W., Pantofaru, C., & Savarese, S. (2014, September). Discovering groups of people in images. In European conference on computer vision (pp. 417-433). Springer, Cham.
  • Akar, A., & Ikizler-Cinbis, N. (2019, September). Mask Guided Fusion for Group Activity Recognition in Images. In International Conference on Image Analysis and Processing (pp. 282-291). Springer, Cham.
  • Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems (pp. 1097-1105).
  • Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
  • Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9).
  • He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
  • Li, X., & Choo Chuah, M. (2017). SBGAR: semantics based group activity recognition. In Proceedings of the IEEE international conference on computer vision (pp. 2876-2885).

Grup Aktiviteleri Üzerinde Görüntü Tanıma Yöntemi Analizi

Year 2020, Ejosat Special Issue 2020 (HORA), 68 - 72, 15.08.2020
https://doi.org/10.31590/ejosat.779063

Abstract

Sabit resimler üzerinde grup aktivitesi tanıma oldukça zorlayıcı bir problemdir. Resimler üzerinde ön ve arka plan ayrımını yapmak, uzamsal ve zamansal bilginin olmaması nedeniyle, bu problemi video üzerinde grup aktivite tanıma problemine göre daha zor kılmaktadır. Bu çalışmada grup aktivite tanıma için oluşturulmuş olan Volleyball video veri kümesi üzerinde, sabit resim tanıma tabanlı bir yöntemi incelemekteyiz. Kullanılan veri kümesindeki her bir video için bulunan hedef video karesine ek olarak, hedef karenin önceki ve sonraki karelerinden elde edilen ortalama görüntülerin resimler üzerinde zamansal bilgiye yaptığı katkının analizi incelenmektedir. Video grup aktivite tanıma problemlerinde sıkça kullanılan zamansal bilgi, ortalama resim kareleri üzerinden elde edilmekte ve resim tanıma yönteminin eğitim aşamasında kullanılmaktadır. Deney sonuçlarından anlaşıldığı üzere, önerilen yöntemimiz son teknoloji zamansal bilgiyi kullanan video tabanlı çalışmalar ile karşılaştırılabilecek sonuçlar alabilmektedir.

References

  • Ibrahim, M. S., Muralidharan, S., Deng, Z., Vahdat, A., & Mori, G. (2016). A hierarchical deep temporal model for group activity recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1971-1980).
  • Shu, T., Todorovic, S., & Zhu, S. C. (2017). CERN: confidence-energy recurrent network for group activity recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 5523-5531).
  • Qi, M., Qin, J., Li, A., Wang, Y., Luo, J., & Van Gool, L. (2018). stagnet: An attentive semantic RNN for group activity recognition. In Proceedings of the European Conference on Computer Vision (ECCV) (pp. 101-117).
  • Lu, L., Di, H., Lu, Y., Zhang, L., & Wang, S. (2018). A two-level attention-based interaction model for multi-person activity recognition. Neurocomputing, 322, 195-205.
  • Lu, L., Lu, Y., Yu, R., Di, H., Zhang, L., & Wang, S. (2019). GAIM: Graph Attention Interaction Model for Collective Activity Recognition. IEEE Transactions on Multimedia, 22(2), 524-539.
  • Zalluhoglu, C., & Ikizler-Cinbis, N. (2019). Region based multi-stream convolutional neural networks for collective activity recognition. Journal of Visual Communication and Image Representation, 60, 170-179.
  • Choi, W., Chao, Y. W., Pantofaru, C., & Savarese, S. (2014, September). Discovering groups of people in images. In European conference on computer vision (pp. 417-433). Springer, Cham.
  • Akar, A., & Ikizler-Cinbis, N. (2019, September). Mask Guided Fusion for Group Activity Recognition in Images. In International Conference on Image Analysis and Processing (pp. 282-291). Springer, Cham.
  • Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems (pp. 1097-1105).
  • Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
  • Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9).
  • He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
  • Li, X., & Choo Chuah, M. (2017). SBGAR: semantics based group activity recognition. In Proceedings of the IEEE international conference on computer vision (pp. 2876-2885).
There are 13 citations in total.

Details

Primary Language English
Subjects Engineering
Journal Section Articles
Authors

Cemil Zalluhoğlu 0000-0001-8716-6297

Publication Date August 15, 2020
Published in Issue Year 2020 Ejosat Special Issue 2020 (HORA)

Cite

APA Zalluhoğlu, C. (2020). Analysis of an Image Recognition Method on Group Activities. Avrupa Bilim Ve Teknoloji Dergisi68-72. https://doi.org/10.31590/ejosat.779063