Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
698 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...We use the VGG-16 architecture [41] for its state-of-the-art performance [39]....
[...]
697 citations
Cites background from "Very Deep Convolutional Networks fo..."
..., layer relu5 4) of the pre-trained 19 layers VGG network [74]....
[...]
691 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...We choose several popular CNN architectures as our teacher/student models, namely, AlexNet [27], AlexNet with Tucker Decomposition [26], VGG16 [37] and VGGM [4]....
[...]
...In the first set of experiments, we use a smaller network (that is, less parameters) as the student and use a larger one for the teacher (for example, AlexNet as student and VGG16 as teacher)....
[...]
...We choose VGG16 as the teacher model and Tucker as our student model for all the experiments in this section....
[...]
...Typically, when evaluating efficiency, we get 3 times faster from VGG16 as teacher to AlexNet as student on KITTI dataset....
[...]
...Table 5 compares the accuracy of Tucker model learned with VGG16 on the trainval and testing split of the PASCAL and COCO datasets....
[...]
689 citations
Cites background from "Very Deep Convolutional Networks fo..."
...Theoretically, other convolutional networks like GoogleNet[19], VGG ILSVRC 19 layers[17] can also be embedded in our framework....
[...]
685 citations
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations