Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
548 citations
546 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...For this experiment, our base architecture is the VGG-16 architecture, initializing from weights pretrained on ImageNet [29]....
[...]
...For this experiment, our base architecture is the VGG-16 architecture, initializing from weights pretrained on ImageNet [24]....
[...]
546 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...ful methods for learning feature representations automatically from data [3]–[5]....
[...]
545 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...In Table 5, we present the adaptation result on the task “GTA5 to Cityscapes” with ResNet101 and VGG16....
[...]
...We compare the results between our method and the state-of-the-art method with two different backbone networks: ResNet101 and VGG16 respectively....
[...]
...For “SYNTHIA to Cityscapes” with 13 categories for ResNet101 and 16 categories for VGG16, the upper bounds are 71.7 and 59.5....
[...]
...We can observe the role of backbone in all domain adaptation methods, namely ResNet101 achieves a much better result than VGG16....
[...]
...For FCN-8s with VGG16, we use Adam as the optimizer with momentum as 0.9 and 0.99....
[...]
544 citations
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations