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 |
---|---|---|
General Classification | 30 | 13.70% |
Image Classification | 29 | 13.24% |
Object Detection | 18 | 8.22% |
Computer Vision | 12 | 5.48% |
Quantization | 11 | 5.02% |
Object Recognition | 8 | 3.65% |
Autonomous Driving | 5 | 2.28% |
Domain Adaptation | 5 | 2.28% |
Image Generation | 3 | 1.37% |
Component | Type |
|
---|---|---|
![]() |
Convolutions | |
![]() |
Convolutions | |
![]() |
Pooling Operations |