Selective Kernel

Introduced by Li et al. in Selective Kernel Networks

A Selective Kernel unit is a bottleneck block consisting of a sequence of 1×1 convolution, SK convolution and 1×1 convolution. It was proposed as part of the SKNet CNN architecture. In general, all the large kernel convolutions in the original bottleneck blocks in ResNeXt are replaced by the proposed SK convolutions, enabling the network to choose appropriate receptive field sizes in an adaptive manner.

In SK units, there are three important hyper-parameters which determine the final settings of SK convolutions: the number of paths $M$ that determines the number of choices of different kernels to be aggregated, the group number $G$ that controls the cardinality of each path, and the reduction ratio $r$ that controls the number of parameters in the fuse operator. One typical setting of SK convolutions is $\text{SK}\left[M, G, r\right]$ to be $\text{SK}\left[2, 32, 16\right]$.

Source: Selective Kernel Networks

Read Paper See Code

Papers

Paper	Code	Results	Date	Stars

Tasks

Task	Papers	Share
Object Detection	4	11.76%
Semantic Segmentation	4	11.76%
Image Classification	3	8.82%
Point Cloud Completion	2	5.88%
Image Segmentation	1	2.94%
3D Classification	1	2.94%
Classification	1	2.94%
Object Detection In Aerial Images	1	2.94%
Lesion Segmentation	1	2.94%