Image Model Blocks
Inception Module
Introduced by Szegedy et al. in Going Deeper with ConvolutionsAn 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 Convolutions
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 35 | 9.04% |
| General Classification | 30 | 7.75% |
| Classification | 24 | 6.20% |
| Deep Learning | 21 | 5.43% |
| Object Detection | 21 | 5.43% |
| Quantization | 13 | 3.36% |
| Object | 12 | 3.10% |
| Object Recognition | 8 | 2.07% |
| Autonomous Driving | 7 | 1.81% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
1x1 Convolution
|
Convolutions | |
Convolution
|
Convolutions | |
Max Pooling
|
Pooling Operations |
