Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
666 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...Figure 2 shows a detailed visualisation of our network based on VGG-16 [27], illustrating the encoder half of SegNet....
[...]
...UberNet [16] proposes an image pyramid approach to process images across multiple resolutions, where for each resolution, additional task-specific layers are formed top of the shared VGG-Net [27]....
[...]
662 citations
658 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...We apply dropout on the initial image output from the VGG convolutional neural network (Simonyan & Zisserman, 2014) as well as the input to the answer module, keeping input with probability p = 0.5....
[...]
...Local region feature extraction: To extract features from the image, we use a convolutional neural network (Krizhevsky et al., 2012) based upon the VGG-19 model (Simonyan & Zisserman, 2014)....
[...]
657 citations
Cites background from "Very Deep Convolutional Networks fo..."
...Recently, representations learned using deep neural networks have presented significant improvements over hand-designed features on many computer vision tasks, such as image classification [29, 44, 46, 49], object detection [13, 14, 20, 43], etc....
[...]
657 citations
Cites background or methods from "Very Deep Convolutional Networks fo..."
...Additionally, for UCF-101, we computed flow percepts by extracting flows using the Brox method and training the temporal stream convolutional network as described by Simonyan & Zisserman (2014a)....
[...]
...A Temporal stream convolutional net, similar to the one proposed by Simonyan & Zisserman (2014b), was trained on single frame optical flows and stacks of 10 optical flows....
[...]
...This has led to research in 3D convolutional nets (Ji et al., 2013; Tran et al., 2014), different temporal fusion strategies (Karpathy et al., 2014) and exploring different ways of presenting visual information to convolutional nets (Simonyan & Zisserman, 2014a)....
[...]
...We extracted percepts using the convolutional neural net model of Simonyan & Zisserman (2014b)....
[...]
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations