ADD-Net: Attention Based 3D Dense Network for Action Recognition
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Recent years with the development of artificial intelligence and the success of the deep model, they have been deployed in all fields of computer vision. Action recognition, as an important branch of human perception and computer vision system research, has attracted more and more attention. Action recognition is a challenging task due to the special complexity of human movement, the same movement may exist between multiple individuals. The human action exists as a continuous image frame in the video, so action recognition requires more computational power than processing static images. And the simple use of the CNN network cannot achieve the desired results. Recently, the attention model has achieved good results in computer vision and natural language processing. In particular, for video action classification, after adding the attention model, it is more effective to focus on motion features and improve performance. It intuitively explains which part the model attends to when making a particular decision, which is very helpful in real applications. In this paper, we proposed a 3D dense convolutional network based on attention mechanism（ADD-Net）, recognition of human motion behavior in the video.
Abstract
Recent years with the development of artificial intelligence and the success of the deep model, they have been deployed in all fields of computer vision. Action recognition, as an important branch of human perception and computer vision system research, has attracted more and more attention. Action recognition is a challenging task due to the special complexity of human movement, the same movement may exist between multiple individuals. The human action exists as a continuous image frame in the video, so action recognition requires more computational power than processing static images. And the simple use of the CNN network cannot achieve the desired results. Recently, the attention model has achieved good results in computer vision and natural language processing. In particular, for video action classification, after adding the attention model, it is more effective to focus on motion features and improve performance. It intuitively explains which part the model attends to when making a particular decision, which is very helpful in real applications. In this paper, we proposed a 3D dense convolutional network based on attention mechanism（ADD-Net）, recognition of human motion behavior in the video.
참고문헌 (38)
- K. Simonyan, "Two-stream convolutional networks for action recognition in videos", NIPS (2014)
- H. Kuehne, "High Performance Computing in Science and Engineering" (2013)
- K. Soomro, "Shah. Ucf101: A dataset of 101 human actions classes from videos in the wild", arXiv:1212.0402, ~ (2012)
- H. Bilen, "Dynamic image networks for action recognition", Proc. CVPR (2016)
- Christoph Feichtenhofer, "Convolutional two-stream network fusion for video action recognition", Proc. CVPR (2016)
- A. Krizhevsky, "ImageNet classification with deep convolu-tional neural networks", NIPS (2012)
- K. Simonyan, "Two-stream convolutional networks for action recognition in videos", NIPS (2014)
- Heng Wang, "Action recognition with improved trajectories. In P with improved trajectories", Proc. ICCV (2013)
- J. Ba, "Multiple object recognition with visual attention", Proc. Int. Conf. Learn. Represent (2015)
- S. Sharma, "Action recognition using visual attention", Proc. Int. Conf. Learn. Represent. Workshop (2016)
- A. Diba, "Efficient two stream motion and appearance 3d cnns for video clas-sification", ECCV Workshops (2016)
- A. Diba, "Deep temporal linear encoding networks", CVPR (2017)
- C. Feichtenhofer, "Spatio-temporal residual networks for video ac-tion recognition", Advances in Neural Information Processing Systems, pp.3468~3476 (2016)
- C. Feichtenhofer, "Convolutio nal two-stream network fusion for video action recognition", CVPR (2016)
- L. Wang, "Temporal segment networks: Towards good practices for deep action recognition", ECCV, (2016) (2016)
- J. Yue-Hei Ng, "Beyond short snippets: Deep networks for video classification", CVPR (2015)
- H. Wang, "Action recognition with improved trajectories", ICCV (2013)
- D. Tran, "Convnet architecture search for spatio-temporal feature learning", arXiv:1708.05038, ~ (2017)
- L. Sun, "Human action recognition using factorized spa-tio-temporal convolutional networks", ICCV (2015)
- A. Karpathy, "Large-scale video classification with convolutional neural networks", CVPR (2014)
- G. Huang, "Densely connected convolu-tional networks", CVPR (2017)
- N. Dalal, "Human detection using oriented histograms of flow and appearance", ECCV (2006)
- J. Deng, "Imagenet: A large-scale hierar-chical image database", CVPR (2009)
- Yang, H., "Convolutional Neural Networks for Human Action Recognition", IEEE Trans. Pattern Anal. Mach. Intell (2012)
- Tran, D., "Learning spatiotemporal fea-tures with 3d convolutional networks", Proceedings of the 2015International Conference on Computer Vision (2015)
- Zinkevich, M., "Parallelized stochastic gradient descent", Proceedings of the Advances in Neural Information Processing Systems, pp.2595~2603 (2010)
- Sutskever, I., "On the importance of initialization and mo-mentum in deep learning", Proceedings of the International Conference on Machine Learning, pp.1139~1147 (2013)
- J. Carreira, "Quo vadis, action recognition? a new model and the kinetics dataset", CVPR (2017)
- A. Klaser, "A spatio-temporal descriptor based on 3d-gradients", BMVC (2008)
- K. He, "Deep residual learning for image recognition", CVPR (2016)
- K. Xu, "Show, attend and tell: Neural image caption generation with visual attention", Proc. ICML, pp.2048~2057 (2015)
- C. Zhu, "Structured attentions for visual question an-swering", Proc. ICCV, 3, pp.1300~1309 (2017)
- W. Zou, "A comparable study of modeling units for end-to-end mandarin speech recognition", arXiv preprint arXiv:1805.03832, ~ (2018)
- S. Kim, "Joint ctc-attention based endto-end speech recognition using multi-task learning", Acoustics, Speech and Signal Processing (ICASSP), 2017 IEEE International Conference on, pp.4835~4839 (2017), IEEE
- Rush, A. M., "A Neural Attention Model for Abstractive Sentence Summariza-tion", EMNLP (2015)
- "2016", ~
- Hu, J, "Squeeze-and-excitation networks", arXiv preprint arXiv:1709.01507, ~ (2017)
- Wang, F, "Residual atten-tion network for image classification", arXiv preprint arXiv:1704.06904, ~ (2017)