arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1701.01077cs.CV
收藏

将传感器数据转换到图像域以进行深度学习-脚步检测的应用

Transforming Sensor Data to the Image Domain for Deep Learning - an Application to Footstep Detection

TU Kaiserslautern(凯撒斯劳滕大学);MindGarage, TU Kaiserslautern(MindGarage,凯撒斯劳滕大学);DFKI(德国人工智能研究中心);DIVA, University of Fribourg(弗里堡大学DIVA)

Monit Shah Singh, Vinaychandran Pondenkandath, Bo Zhou, Paul Lukowicz, Marcus Liwicki

摘要:

卷积神经网络(CNN)已经成为各种计算机视觉任务中的最先进技术,但对于大多数传感器数据,特别是在普适和可穿戴计算中,它们仍然是不成熟的。其主要原因是有限数量的带注释的训练数据。在本文中,我们提出了通过将传感器模态转换到视觉域来利用预训练的深度CNN对二维传感器数据的区分能力的想法。通过提出的三种策略,将二维传感器输出转换为压力分布图像。然后,我们利用预训练的CNN对转换后的图像数据进行迁移学习。我们评估我们的方法上的步态数据集的地板表面压力映射。我们获得了87.66%的分类准确率,这比传统的机器学习方法超过10%。

英文摘要:

Convolutional Neural Networks (CNNs) have become the state-of-the-art in various computer vision tasks, but they are still premature for most sensor data, especially in pervasive and wearable computing. A major reason for this is the limited amount of annotated training data. In this paper, we propose the idea of leveraging the discriminative power of pre-trained deep CNNs on 2-dimensional sensor data by transforming the sensor modality to the visual domain. By three proposed strategies, 2D sensor output is converted into pressure distribution imageries. Then we utilize a pre-trained CNN for transfer learning on the converted imagery data. We evaluate our method on a gait dataset of floor surface pressure mapping. We obtain a classification accuracy of 87.66%, which outperforms the conventional machine learning methods by over 10%.

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。

↑