________________________________>
| ADD --> f(x) + x
x-----> 1x1 + 3x3 + 1x1 卷积 ----->
1. 步长为1结合x shortcut
___________________________________>
| ADD --> f(x) + x
x-----> 1x1 + 3x3DW + 1x1 卷积 ----->
“扩张”→“卷积提特征”→ “压缩”
ResNet是:压缩”→“卷积提特征”→“扩张”,MobileNetV2则是Inverted residuals,即:“扩张”→“卷积提特征”→ “压缩”
2. 步长为2时不结合x
x-----> 1x1 + 3x3DW(步长为2) + 1x1 卷积 -----> 输出
普通点卷积时间还是较长
___________________________________________________________>
| ADD --> f(x) + x
x-----> 1x1分组点卷积 + 通道重排 + 3x3DW + 1x1分组点卷积 ----->
_____________________3*3AvgPool_________________________________>
| concat --> f(x) 链接 x
x-----> 1x1分组点卷积 + 通道重排 + 3x3DW步长2 + 1x1分组点卷积 ----->
group 1 | group 2 | group 3 |
---|---|---|
1 2 | 3 4 | 5 6 |
Each nubmer represents a channel of the feature map
1 2
3 4
5 6
1 3 5
2 4 6
group 1 | group 2 | group 3 |
---|---|---|
1 3 | 5 2 | 4 6 |