While usage of convolutional neural networks (CNN) is widely prevalent, methods proposed so far always have considered homogeneous kernels for this task. In this paper, we propose a new type of convolution operation using heterogeneous kernels. The proposed Heterogeneous Kernel-Based Convolution (HetConv) reduces the computation (FLOPs) and the number of parameters as compared to standard convolution operation while it maintains representational efficiency. To show the effectiveness of our proposed convolution, we present extensive experimental results on the standard CNN architectures such as VGG, ResNet, Faster-RCNN, MobileNet, and SSD. We observe that after replacing the standard convolutional filters in these architectures with our proposed HetConv filters, we achieve 1.5 ×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document} to 8 ×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document} FLOPs based improvement in speed while it maintains (sometimes improves) the accuracy. We also compare our proposed convolution with group/depth wise convolution and show that it achieves more FLOPs reduction with significantly higher accuracy. Moreover, we demonstrate the efficacy of HetConv based CNN by showing that it also generalizes on object detection and is not constrained to image classification tasks. We also empirically show that the proposed HetConv convolution is more robust towards the over-fitting problem as compared to standard convolution.

中文翻译：

---

HetConv：超越深度 CNN 的同构卷积核

虽然卷积神经网络 (CNN) 的使用广泛流行，但迄今为止提出的方法总是考虑使用同构内核来完成这项任务。在本文中，我们提出了一种使用异构内核的新型卷积运算。与标准卷积操作相比，所提出的基于异构内核的卷积 (HetConv) 减少了计算 (FLOP) 和参数数量，同时保持了表示效率。为了展示我们提出的卷积的有效性，我们在标准 CNN 架构（如 VGG、ResNet、Faster-RCNN、MobileNet 和 SSD）上展示了大量实验结果。我们观察到，在用我们提出的 HetConv 过滤器替换这些架构中的标准卷积过滤器后，我们实现了 1。5 ×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin }{-69pt} \begin{document}$$\times $$\end{document} 到 8 ×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{ amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document} FLOPs 基于速度的改进保持（有时提高）准确性。我们还将我们提出的卷积与组/深度明智的卷积进行了比较，并表明它以更高的精度实现了更多的 FLOPs 减少。而且，我们展示了基于 HetConv 的 CNN 的功效，展示了它也可以泛化对象检测并且不限于图像分类任务。我们还凭经验表明，与标准卷积相比，所提出的 HetConv 卷积对过拟合问题更稳健。