arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1902.03227cs.CVeess.IV
收藏

深度神经网络中的最小图像:自然图像中的脆弱物体识别

Minimal Images in Deep Neural Networks: Fragile Object Recognition in Natural Images

Center for Brains, Minds, and Machines (CBMM)(脑、心智与机器中心(CBMM));Massachusetts Institute of Technology(麻省理工学院);Children’s Hospital, Harvard Medical School(哈佛医学院儿童医院)

Sanjana Srivastava, Guy Ben-Yosef, Xavier Boix

摘要:

当物体没有完全显示出来时,人类识别物体的能力就会受到损害。“最小图像”是图像中人类可识别的最小区域。Ullman等人2016年的研究表明,对最小图像可见区域的位置和大小进行轻微修改会导致人类识别准确性急剧下降。在本文中,我们证明了由于可见区域的变化而导致的准确性下降是人类和现有最先进的深度神经网络(DNN)之间的常见现象,并且在DNN中更为突出。我们发现,在许多情况下,DNN对一个区域的分类是正确的,而对另一个区域的分类是错误的,尽管它们只相差一行或一列像素,而且通常比人类平均最小图像尺寸大。我们表明,这种现象是独立于以前的作品,已报告缺乏不变性的DNN中的对象位置的微小修改。因此,我们的研究结果揭示了DNN的一种新的故障模式,这种模式对人类的影响要小得多。他们揭示了DNN对自然图像的识别能力是多么脆弱,即使没有引入对抗模式。将自然图像中DNN的鲁棒性带到人类水平仍然是社区面临的一个公开挑战。

英文摘要:

The human ability to recognize objects is impaired when the object is not shown in full. "Minimal images" are the smallest regions of an image that remain recognizable for humans. Ullman et al. 2016 show that a slight modification of the location and size of the visible region of the minimal image produces a sharp drop in human recognition accuracy. In this paper, we demonstrate that such drops in accuracy due to changes of the visible region are a common phenomenon between humans and existing state-of-the-art deep neural networks (DNNs), and are much more prominent in DNNs. We found many cases where DNNs classified one region correctly and the other incorrectly, though they only differed by one row or column of pixels, and were often bigger than the average human minimal image size. We show that this phenomenon is independent from previous works that have reported lack of invariance to minor modifications in object location in DNNs. Our results thus reveal a new failure mode of DNNs that also affects humans to a much lesser degree. They expose how fragile DNN recognition ability is for natural images even without adversarial patterns being introduced. Bringing the robustness of DNNs in natural images to the human level remains an open challenge for the community.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑