Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
This addresses a crucial problem in visual tasks for researchers and practitioners by providing insights into network design, though it is incremental as it builds on existing concepts.
The paper tackled the issue of receptive field size in deep convolutional neural networks, showing that the effective receptive field has a Gaussian distribution and occupies only a fraction of the theoretical size, with analysis leading to suggestions for improvement.
We study characteristics of receptive fields of units in deep convolutional networks. The receptive field size is a crucial issue in many visual tasks, as the output must respond to large enough areas in the image to capture information about large objects. We introduce the notion of an effective receptive field, and show that it both has a Gaussian distribution and only occupies a fraction of the full theoretical receptive field. We analyze the effective receptive field in several architecture designs, and the effect of nonlinear activations, dropout, sub-sampling and skip connections on it. This leads to suggestions for ways to address its tendency to be too small.