Convolution is spatially-symmetric, i.e., the visual features are independent of its position in the image, which limits its ability to utilize contextual cues for visual recognition. This paper addresses this issue by introducing a recalibration process, which refers to the surrounding region of each neuron, computes an importance value and multiplies it to the original neural response. Our approach is named multi-scale spatially-asymmetric recalibration (MS-SAR), which extracts visual cues from surrounding regions at multiple scales, and designs a weighting scheme which is asymmetric in the spatial domain. MS-SAR is implemented in an efficient way, so that only small fractions of extra parameters and computations are required. We apply MS-SAR to several popular building blocks, including the residual block and the densely-connected block, and demonstrate its superior performance in both CIFAR and ILSVRC2012 classification tasks.
@article{arxiv.1804.00787,
title = {Multi-Scale Spatially-Asymmetric Recalibration for Image Classification},
author = {Yan Wang and Lingxi Xie and Siyuan Qiao and Ya Zhang and Wenjun Zhang and Alan L. Yuille},
journal= {arXiv preprint arXiv:1804.00787},
year = {2018}
}