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 1703.09844cs.LG
收藏

基于多尺度稠密网络的资源高效图像分类

Multi-Scale Dense Networks for Resource Efficient Image Classification

Cornell University(康奈尔大学);Fudan University(复旦大学);Tsinghua University(清华大学);Facebook AI Research(Facebook人工智能研究(Facebook AI Research))

Gao Huang, Danlu Chen, Tianhong Li, Felix Wu, Laurens van der Maaten, Kilian Q. Weinberger

摘要:

在本文中,我们研究图像分类的计算资源限制在测试时间。两个这样的设置是:1.随时分类,其中网络对测试示例的预测逐步更新,便于在任何时间输出预测;以及2.预算批量分类,其中固定量的计算可用于对一组示例进行分类,这些示例可能在“更容易”和“更难”的输入之间不均匀地花费。与大多数以前的工作,如流行的Viola和Jones算法相比,我们的方法是基于卷积神经网络。我们训练多个分类器,不同的资源需求,我们自适应地应用在测试时间。为了最大限度地重用分类器之间的计算,我们将它们作为早期退出合并到单个深度卷积神经网络中,并将它们与密集连接进行互连。为了促进早期的高质量分类,我们使用了一个二维的多尺度网络架构,在整个网络中保持粗和细级别的功能。三个图像分类任务的实验表明,我们的框架大大提高了现有的国家的最先进的在这两种设置。

英文摘要:

In this paper we investigate image classification with computational resource limits at test time. Two such settings are: 1. anytime classification, where the network's prediction for a test example is progressively updated, facilitating the output of a prediction at any time; and 2. budgeted batch classification, where a fixed amount of computation is available to classify a set of examples that can be spent unevenly across "easier" and "harder" inputs. In contrast to most prior work, such as the popular Viola and Jones algorithm, our approach is based on convolutional neural networks. We train multiple classifiers with varying resource demands, which we adaptively apply during test time. To maximally re-use computation between the classifiers, we incorporate them as early-exits into a single deep convolutional neural network and inter-connect them with dense connectivity. To facilitate high quality classification early on, we use a two-dimensional multi-scale network architecture that maintains coarse and fine level features all-throughout the network. Experiments on three image-classification tasks demonstrate that our framework substantially improves the existing state-of-the-art in both settings.

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。

↑