Circular Kernel

We provide code and our dataset for the Paper:

Integrating Large Circular Kernels into CNNs through Neural Architecture Search
Kun He, Chao Li, Yixiao Yang, Gao Huang, John E. Hopcroft\

Introduction

The square convolution kernel has been regarded as the standard and core unit of Convolutional Neural Networks (CNNs) since the first recognized CNN of LeNet proposed in 1989, and especially after AlexNet won the ILSVRC (ImageNet Large Scale Visual Recognition Competition) in 2012. Since then, various variants of convolution kernels have been proposed, including separable convolution, dilated convolution, deformable convolution, \etc. Inspired by the fact that the retinal ganglion cells in the biological visual system have approximately concentric receptive fields (RFs), we propose the concept of circular kernels for the convolution operation. A $K \times K$ circular kernel is defined as a kernel that evenly samples K² pixels on the concentric circles to form a circular receptive field.

Getting Started

Environment Setup

To set up the enviroment you can easily run the following command:

conda create -n CK python=3.6
conda activate CK
pip install -r requirements.txt

Data Preparation

You need to first download the ImageNet-2012 to the folder ./data/imagenet and move the validation set to the subfolder ./data/imagenet/val. To move the validation set, you cloud use the following script: https://raw.githubusercontent.com/soumith/imagenetloader.torch/master/valprep.sh

The directory structure is the standard layout as following.

/path/to/imagenet/
  train/
    class1/
      img1.jpeg
    class2/
      img2.jpeg
  val/
    class1/
      img3.jpeg
    class/2
      img4.jpeg

Run

Search on CIFAR10

python train_search.py --operation PRIMITIVES_circle_square \\

Search on ImageNet

Data preparation: 10% and 2.5% images need to be random sampled prior from earch class of trainingset as train and val, respectively. The sampled data is save into ./imagenet_search. Note that not to use torch.utils.data.sampler.SubsetRandomSampler for data sampling as imagenet is too large.

python train_search_imagenet.py 
		--operation PRIMITIVES_circle_square \\
		--tmp_data_dir /path/to/your/sampled/data \\

Evaluation on CIFAR10:

python train.py \\
       --auxiliary \\
       --cutout \\
       --arch  PC_DARTS_Circle_cifar 

Evaluation on ImageNet (mobile setting):

python train_imagenet.py \\
       --tmp_data_dir /path/to/your/data \\
       --auxiliary \\
       --arch PC_DARTS_Circle_image

Pretrained models

Coming soon!

Reference