We focus on building robustness in the convolutions of neural visual classifiers, especially against natural perturbations like elastic deformations, occlusions and Gaussian noise. Existing CNNs show outstanding performance on clean images, but fail to tackle naturally occurring perturbations. In this paper, we start from elastic perturbations, which approximate (local) view-point changes of the object. We present elastically-augmented convolutions (EAConv) by parameterizing filters as a combination of fixed elastically-perturbed bases functions and trainable weights for the purpose of integrating unseen viewpoints in the CNN. We show on CIFAR-10 and STL-10 datasets that the general robustness of our method on unseen occlusion, zoom, rotation, image cut and Gaussian perturbations improves, while significantly improving the performance on clean images without any data augmentation.
翻译:在神经视觉分类器的演进中,我们的重点是建设坚固的神经视觉分级器,特别是防止自然扰动,例如弹性变形、闭塞和高斯噪音。现有的CNN显示清洁图像的出色表现,但未能处理自然发生的扰动。在本文中,我们从弹性扰动开始,这大约是物体的(当地)视图点变化。我们通过将过滤器参数化,将固定的弹性透视基础功能和可训练重量结合起来,将隐形观点纳入CNN,从而显著改进清洁图像的性能,而没有任何数据增强。我们在CIFAR-10和STL-10数据集中显示,我们关于隐形封闭、缩放、旋转、图像切割和高斯扰动的方法的总体稳健性有所改善,同时大大改进清洁图像的性能,而没有任何数据增强。