An Inception Module is an image model block that aims to approximate an optimal local sparse structure in a CNN. Put simply, it allows for us to use multiple types of filter size, instead of being restricted to a single filter size, in a single image block, which we then concatenate and pass onto the next layer.
Source: Going Deeper with ConvolutionsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 35 | 11.67% |
General Classification | 30 | 10.00% |
Classification | 24 | 8.00% |
Object Detection | 20 | 6.67% |
Quantization | 13 | 4.33% |
Object Recognition | 8 | 2.67% |
Autonomous Driving | 6 | 2.00% |
Domain Adaptation | 5 | 1.67% |
Clustering | 5 | 1.67% |
Component | Type |
|
---|---|---|
1x1 Convolution
|
Convolutions | |
Convolution
|
Convolutions | |
Max Pooling
|
Pooling Operations |