GMConv：调制卷积核的有效感受野
GMConv: Modulating Effective Receptive Fields for Convolutional Kernels
AI总结：
针对方形卷积核固定感受野与实际高斯型 ERF 不匹配的问题，提出可即插即用、端到端训练的 GMConv，并在分类和检测任务上验证其稳定增益。
摘要：
在卷积神经网络中，卷积传统上使用具有固定 N×N 感受野（RF）的方形核来执行。然而，对网络最重要的是有效感受野（ERF），它表示输入像素对输出像素产生贡献的程度。受 ERF 通常呈现高斯分布这一特性启发，我们在本工作中提出了一种高斯掩码卷积核（GMConv）。具体而言，GMConv 利用高斯函数生成一个同心对称掩码，并将其放置在卷积核之上以细化 RF。我们的 GMConv 可以直接替换现有 CNN 中的标准卷积，并且能够通过标准反向传播轻松进行端到端训练。我们通过在图像分类和目标检测任务上的大量实验评估了该方法。在多个任务和标准基础模型上，我们的方法相比标准卷积表现更优。例如，将 GMConv 用于 AlexNet 和 ResNet-50 时，ImageNet 分类的 top-1 准确率分别提升了 0.98% 和 0.85%。
英文摘要：
In convolutional neural networks, the convolutions are conventionally performed using a square kernel with a fixed N $\times$ N receptive field (RF). However, what matters most to the network is the effective receptive field (ERF) that indicates the extent with which input pixels contribute to an output pixel. Inspired by the property that ERFs typically exhibit a Gaussian distribution, we propose a Gaussian Mask convolutional kernel (GMConv) in this work. Specifically, GMConv utilizes the Gaussian function to generate a concentric symmetry mask that is placed over the kernel to refine the RF. Our GMConv can directly replace the standard convolutions in existing CNNs and can be easily trained end-to-end by standard back-propagation. We evaluate our approach through extensive experiments on image classification and object detection tasks. Over several tasks and standard base models, our approach compares favorably against the standard convolution. For instance, using GMConv for AlexNet and ResNet-50, the top-1 accuracy on ImageNet classification is boosted by 0.98% and 0.85%, respectively.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
