Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
343 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...To enforce a fine-grained control using the input sketch, we choose the ReLU2-2 layer of the VGG-19 net [44] to compute the feature loss, since higher level feature representations tend to encourage the network to ignore important details such as the exact locations of the pupils....
[...]
343 citations
343 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...How does the architecture of FA influence knockoff performance? We study the influence using two choices of the blackbox FV architecture: Resnet-34 [13] and VGG16 [32]....
[...]
...We study the influence using two choices of the blackbox FV architecture: Resnet-34 [13] and VGG16 [32]....
[...]
...Keeping these fixed, we vary architecture of the knockoff FA by choosing from: Alexnet [20], VGG-16 [32], Resnet-{18, 34, 50, 101} [13], and Densenet-161 [15]....
[...]
...From Figure 10, we observe: (i) performance of the knockoff ordered by model complexity: Alexnet (lowest performance) is at one end of the spectrum while significantly more complex Resnet-101/Densenet-161 are at the other; (ii) performance transfers across model families: Resnet-34 achieves similar performance when stealing VGG-16 and vice versa; (iii) complexity helps: selecting a more complex model architecture of the knockoff is beneficial....
[...]
343 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...The VGG-16 model [41] is simply adapted to be our deep hash model in this work....
[...]
[...]
342 citations
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations