Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
1,561 citations
1,558 citations
Cites methods from "Very Deep Convolutional Networks fo..."
...A typical algorithm in the image processing domain starts with multiple convolutional layers that first extract basic feature maps, followed by more complex feature maps....
[...]
...Given the relative scarcity of NFUs, DaDianNao adopts the following approach to maximize performance....
[...]
1,537 citations
1,535 citations
Cites background from "Very Deep Convolutional Networks fo..."
...[15, 16]) and the accessible computation power is increasing, the size of the datasets for training and evaluation is not increasing by much, but rather lagging behind and hindering further progress in large-scale visual recognition....
[...]
1,527 citations
Cites background or methods from "Very Deep Convolutional Networks fo..."
...State-of-the-art image features are generally extracted by deep Convolutional Neural Networks (CNNs) [8, 25, 32]....
[...]
...This model is generic and thus can be applied to any layer in any CNN architecture such as popular VGG [25] and ResNet [8]....
[...]
...Different from existing popular modulating strategy that sums up all visual features based on attention weights [34], function f(·) applies element-wise multiplication....
[...]
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations