arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.04127cs.LG
收藏

CNN移位不变性是什么样的?可视化研究

What Does CNN Shift Invariance Look Like? A Visualization Study

Columbia University(哥伦比亚大学);The University of Texas at Austin(德克萨斯大学奥斯汀分校)

Jake Lee, Junfeng Yang, Zhangyang Wang

摘要:

利用卷积神经网络(CNN)进行特征提取是一种表示用于机器学习任务的图像的流行方法。这些表示寻求捕获全局图像内容,并且理想地应该独立于几何变换。我们专注于测量和可视化从流行的现成的CNN模型中提取的特征的移位不变性。我们提出了三个实验的结果比较表示数百万的图像与彻底移位的对象,检查局部不变性(在几个像素内)和全局不变性(整个图像帧)。我们的结论是,从流行的网络中提取的特征不是全局不变的,并且偏差和伪影存在于该方差内。此外,我们确定抗锯齿模型显着提高局部不变性,但不影响全局不变性。最后,我们提供了一个用于实验再现的代码库,以及一个与我们的结果进行交互的网站https://jakehlee.github.io/visualize-invariance。

英文摘要:

Feature extraction with convolutional neural networks (CNNs) is a popular method to represent images for machine learning tasks. These representations seek to capture global image content, and ideally should be independent of geometric transformations. We focus on measuring and visualizing the shift invariance of extracted features from popular off-the-shelf CNN models. We present the results of three experiments comparing representations of millions of images with exhaustively shifted objects, examining both local invariance (within a few pixels) and global invariance (across the image frame). We conclude that features extracted from popular networks are not globally invariant, and that biases and artifacts exist within this variance. Additionally, we determine that anti-aliased models significantly improve local invariance but do not impact global invariance. Finally, we provide a code repository for experiment reproduction, as well as a website to interact with our results at https://jakehlee.github.io/visualize-invariance.

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。

↑