Image Model Blocks
ShuffleNet Block
Introduced by Zhang et al. in ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile DevicesA ShuffleNet Block is an image model block that utilises a channel shuffle operation, along with depthwise convolutions, for an efficient architectural design. It was proposed as part of the ShuffleNet architecture. The starting point is the Residual Block unit from ResNets, which is then modified with a pointwise group convolution and a channel shuffle operation.
Source: ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 5 | 8.77% |
| Object Detection | 5 | 8.77% |
| Semantic Segmentation | 5 | 8.77% |
| Model Compression | 4 | 7.02% |
| Real-Time Semantic Segmentation | 3 | 5.26% |
| Computational Efficiency | 2 | 3.51% |
| Network Pruning | 2 | 3.51% |
| Rain Removal | 2 | 3.51% |
| Single Image Deraining | 2 | 3.51% |
Batch Normalization
Channel Shuffle
Depthwise Convolution
Groupwise Point Convolution
Pointwise Convolution
ReLU
Residual Connection
