While standard convolution performs the channelwise and spatial-wise computation in one step, Depthwise Separable Convolution splits the computation into two steps: depthwise convolution applies a single convolutional filter per each input channel and pointwise convolution is used to create a linear combination of the output of the depthwise convolution. The comparison of standard convolution and depthwise separable convolution is shown to the right.
Credit: Depthwise Convolution Is All You Need for Learning Multiple Visual Domains
Source: Xception: Deep Learning With Depthwise Separable ConvolutionsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 72 | 10.47% |
Object Detection | 49 | 7.12% |
Classification | 39 | 5.67% |
Quantization | 33 | 4.80% |
Semantic Segmentation | 30 | 4.36% |
Instance Segmentation | 11 | 1.60% |
Decoder | 10 | 1.45% |
Management | 9 | 1.31% |
Ensemble Learning | 8 | 1.16% |