TY - JOUR
T1 - Deep Insights into Convolutional Networks for Video Recognition
AU - Feichtenhofer, Christoph
AU - Pinz, Axel
AU - Wildes, Richard
AU - Zisserman, Andrew
PY - 2019/10/29
Y1 - 2019/10/29
N2 - As the success of deep models has led to their deployment in all areas of computer vision, it is increasingly important to understand how these representations work and what they are capturing. In this paper, we shed light on deep spatiotemporal representations by visualizing the internal representation of models that have been trained to recognize actions in video. We visualize multiple two-stream architectures to show that local detectors for appearance and motion objects arise to form distributed representations for recognizing human actions. Key observations include the following. First, cross-stream fusion enables the learning of true spatiotemporal features rather than simply separate appearance and motion features. Second, the networks can learn local representations that are highly class specific, but also generic representations that can serve a range of classes. Third, throughout the hierarchy of the network, features become more abstract and show increasing invariance to aspects of the data that are unimportant to desired distinctions (e.g. motion patterns across various speeds). Fourth, visualizations can be used not only to shed light on learned representations, but also to reveal idiosyncrasies of training data and to explain failure cases of the system.
AB - As the success of deep models has led to their deployment in all areas of computer vision, it is increasingly important to understand how these representations work and what they are capturing. In this paper, we shed light on deep spatiotemporal representations by visualizing the internal representation of models that have been trained to recognize actions in video. We visualize multiple two-stream architectures to show that local detectors for appearance and motion objects arise to form distributed representations for recognizing human actions. Key observations include the following. First, cross-stream fusion enables the learning of true spatiotemporal features rather than simply separate appearance and motion features. Second, the networks can learn local representations that are highly class specific, but also generic representations that can serve a range of classes. Third, throughout the hierarchy of the network, features become more abstract and show increasing invariance to aspects of the data that are unimportant to desired distinctions (e.g. motion patterns across various speeds). Fourth, visualizations can be used not only to shed light on learned representations, but also to reveal idiosyncrasies of training data and to explain failure cases of the system.
KW - Computer vision
KW - Machine learning
KW - Deep Learning
KW - Video recognition
KW - Neural network visualization
KW - Action recognition
U2 - 10.1007/s11263-019-01225-w
DO - 10.1007/s11263-019-01225-w
M3 - Article
SN - 0920-5691
JO - International Journal of Computer Vision
JF - International Journal of Computer Vision
ER -