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 2106.06418cs.CVcs.LG
收藏

Scale-Invariant Scale-Channel Networks:将深度网络推广到以前看不见的尺度

Scale-invariant scale-channel networks: Deep networks that generalise to previously unseen scales

KTH Royal Institute of Technology(KTH皇家理工学院)

Ylva Jansson, Tony Lindeberg

摘要:

处理大规模变化的能力对于许多现实世界的视觉任务至关重要。在深度网络中处理尺度的一种简单方法是在一组尺度通道中同时处理多个尺度的图像。然后,原则上,可以通过使用缩放通道之间的权重共享以及来自缩放通道的输出上的最大或平均池化来实现缩放不变性。然而,这种尺度通道网络在显著尺度范围内概括训练集中不存在的尺度的能力以前没有被探索过。在本文中,我们提出了一个系统的研究方法,通过实施不同类型的规模的渠道网络和评估他们的能力,概括到以前看不见的规模。我们开发了一种形式主义的协方差和不变性的尺度通道网络的属性进行分析,并探讨如何不同的设计选择,独特的缩放变换,影响尺度通道网络的整体性能。我们首先表明,两个先前提出的规模通道网络设计不概括以及规模不存在于训练集。我们从理论上解释和实验证明,为什么在这些情况下泛化失败。然后,我们提出了一种新类型的中央凹尺度通道架构,其中尺度通道处理具有降低的分辨率的图像的越来越大的部分。这种新型的规模渠道网络的推广非常好,提供足够的图像分辨率和边界效应的情况下。我们提出的FovMax和FovAvg网络在8的尺度范围内表现几乎相同,在单尺度训练数据上训练时也是如此,并且在小样本制度中从具有大尺度变化的数据集学习时也能提高性能。

英文摘要:

The ability to handle large scale variations is crucial for many real world visual tasks. A straightforward approach for handling scale in a deep network is to process an image at several scales simultaneously in a set of scale channels. Scale invariance can then, in principle, be achieved by using weight sharing between the scale channels together with max or average pooling over the outputs from the scale channels. The ability of such scale channel networks to generalise to scales not present in the training set over significant scale ranges has, however, not previously been explored. In this paper, we present a systematic study of this methodology by implementing different types of scale channel networks and evaluating their ability to generalise to previously unseen scales. We develop a formalism for analysing the covariance and invariance properties of scale channel networks, and explore how different design choices, unique to scaling transformations, affect the overall performance of scale channel networks. We first show that two previously proposed scale channel network designs do not generalise well to scales not present in the training set. We explain theoretically and demonstrate experimentally why generalisation fails in these cases. We then propose a new type of foveated scale channel architecture}, where the scale channels process increasingly larger parts of the image with decreasing resolution. This new type of scale channel network is shown to generalise extremely well, provided sufficient image resolution and the absence of boundary effects. Our proposed FovMax and FovAvg networks perform almost identically over a scale range of 8, also when training on single scale training data, and do also give improved performance when learning from datasets with large scale variations in the small sample regime.

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。

↑