论文标题
在最小可识别的图像贴片上
On the Minimal Recognizable Image Patch
论文作者
论文摘要
与人类的视力相反，共同的识别算法通常在部分遮挡的图像上失败。我们提出，从经验上提出算法限制来表征算法限制，找到最小的可识别贴片（MRP）本身就足以识别图像。一个专门的深网络使我们能够找到给定尺寸的最有用的补丁，并用作实验工具。一项人类视力研究最近对相关（但不同）的最低识别构型（MIRC）[1]进行了表征，为此我们指定了计算类似物（表示CMIRC）。与这些MIRC的尺寸降低相关的人类决策准确性下降是实质而敏锐的。有趣的是，我们指定的计算版本也发现了这种急剧减少。
In contrast to human vision, common recognition algorithms often fail on partially occluded images. We propose characterizing, empirically, the algorithmic limits by finding a minimal recognizable patch (MRP) that is by itself sufficient to recognize the image. A specialized deep network allows us to find the most informative patches of a given size, and serves as an experimental tool. A human vision study recently characterized related (but different) minimally recognizable configurations (MIRCs) [1], for which we specify computational analogues (denoted cMIRCs). The drop in human decision accuracy associated with size reduction of these MIRCs is substantial and sharp. Interestingly, such sharp reductions were also found for the computational versions we specified.
