Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
616 citations
615 citations
Cites background from "Very Deep Convolutional Networks fo..."
...By duplicating an existing architecture, we can initialize the feature extraction part of our model by copying existing VGG model weights that were trained on a very large set of nonmedical image data (i.e., 1.3 million natural images consisting of 1000 different object categories as explained in Simonyan et al.49)....
[...]
...3 million natural images consisting of 1000 different object categories as explained in Simonyan et al.(49))....
[...]
612 citations
Cites background from "Very Deep Convolutional Networks fo..."
...VGG-16 is a combination of five convolutional blocks (13 convolutional layers) and tree fully-connected layers (fc6 to fc8) [23] ....
[...]
611 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...Very deep CNNs [32] were adopted to segment tumors by Pereira et al....
[...]
611 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...To perform the fine-tuning as described in Section 3, we initialize by the convolutional layers of AlexNet [1], VGG16 [23], or ResNet101 [24]....
[...]
...are adopted, such as AlexNet [1], VGG [23], or ResNet [24], while their fully-connected layers are discarded....
[...]
...training the commonly used CNN architectures, such as AlexNet [1], VGG [23], and ResNet [24]....
[...]
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations