milkcat0904/2017-action-recognition-papers

近期要读的2017年顶会的几篇行为识别论文

2017-action-recognition-papers

近期要读的2017年顶会的几篇行为识别论文

1.Compressed Video Action Recognition

Wu C Y, Zaheer M, Hu H, et al. Compressed Video Action Recognition[J]. 
arXiv preprint arXiv:1712.00636, 2017.

• 压缩视频，去除冗余信息

2.What Actions are Needed for Understanding Human Actions in Videos?

Sigurdsson G A, Russakovsky O, Gupta A. What Actions are Needed for Understanding Human Actions in Videos?[J]. 
arXiv preprint arXiv:1708.02696, 2017.（ICCV2017）

• 在charades数据集上，测试了几种传统方法的效果

3.Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding

Sigurdsson G A, Varol G, Wang X, et al. Hollywood in homes: Crowdsourcing data collection for activity understanding[C]//European Conference on Computer Vision. 
Springer International Publishing, 2016: 510-526.（ECCV2016）

• 提出了charades数据集

4.Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset

Carreira J, Zisserman A. Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset[J].
arXiv preprint arXiv:1705.07750, 2017.（CVPR2017）

• 提出了I3D模型，2017年在ucf101,hmdb51两个数据集达到了state of the art 的结果：80.7% on HMDB-51 and 98.0% on UCF-101

5.Attentional Pooling for Action Recognition

Girdhar R, Ramanan D. Attentional pooling for action recognition[C]
//Advances in Neural Information Processing Systems. 2017: 33-44.（NIPS2017）

• 结合注意力系统的行为识别模型

6.Learning Spatio-Temporal Features with 3D Residual Networks for Action Recognition

Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?

Hara K, Kataoka H, Satoh Y. Learning spatio-temporal features with 3D residual networks for action recognition[C]
//Proceedings of the ICCV Workshop on Action, Gesture, and Emotion Recognition. 2017, 2(3): 4.（ICCV2017）

Hara K, Kataoka H, Satoh Y. Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?[J].
arXiv preprint arXiv:1711.09577, 2017.

• 3d-resnet 模型

7.ActionVLAD: Learning spatio-temporal aggregation for action classification

Girdhar R, Ramanan D, Gupta A, et al. ActionVLAD: Learning spatio-temporal aggregation for action classification[J].
arXiv preprint arXiv:1704.02895, 2017.（CVPR2017）

• 将VLAD与action相结合

8.Asynchronous Temporal Fields for Action Recognition

Sigurdsson G A, Divvala S, Farhadi A, et al. Asynchronous Temporal Fields for Action Recognition[J].
arXiv preprint arXiv:1612.06371, 2016.

• 在charades数据集上的行为识别CRF模型