Measuring the similarity of images is a fundamental problem to computer vision for which no universal solution exists. While simple metrics such as the pixel-wise L2-norm have been shown to have significant flaws, they remain popular. One group of recent state-of-the-art metrics that mitigates some of those flaws are Deep Perceptual Similarity (DPS) metrics, where the similarity is evaluated as the distance in the deep features of neural networks. However, DPS metrics themselves have been less thoroughly examined for their benefits and, especially, their flaws. This work investigates the most common DPS metric, where deep features are compared by spatial position, along with metrics comparing the averaged and sorted deep features. The metrics are analyzed in-depth to understand the strengths and weaknesses of the metrics by using images designed specifically to challenge them. This work contributes with new insights into the flaws of DPS, and further suggests improvements to the metrics. An implementation of this work is available online: https://github.com/guspih/deep_perceptual_similarity_analysis/
翻译:测量图像的相似性是计算机视觉中一个根本问题,没有普遍的解决办法。虽然事实证明像素-Wis L2-norm这样的简单度量仪有显著的缺陷,但它们仍然很受欢迎。最近一组最先进的衡量标准减轻了一些缺陷,它们是深层次的感知相似性(DPS)度量仪,其相似性被评为神经网络深层特征的距离。但是,DPS度量仪本身的效益,特别是其缺陷,没有受到更彻底的审查。这项工作调查了最常见的DPS 度量度仪,其中以空间位置比较了深度特征,并比较了平均和分拣的深度特征。通过使用专门设计来挑战这些特征的图像,深入分析了这些尺度的长处和短处。这项工作有助于对DPS的缺陷有新的了解,并进一步建议改进度量度仪。这项工作的落实情况可在网上查阅:https://github.com/guspimic_percepuality_assulation_assyation。