Li Xinyu, Zhang Yanyi, Zhang Jianyu, Chen Shuhong, Gu Yue, Farneth Richard A, Marsic Ivan, Burd Randall S
Rutgers University, Piscataway, New Jersey.
Children's National Medical Center, Washington, District of Columbia.
IPSN. 2017 Apr;2017:297-298. doi: 10.1145/3055031.3055057.
We present a deep learning framework for fast 3D activity localization and tracking in a dynamic and crowded real world setting. Our training approach reverses the traditional activity localization approach, which first estimates the possible location of activities and then predicts their occurrence. Instead, we first trained a deep convolutional neural network for activity recognition using depth video and RFID data as input, and then used the activation maps of the network to locate the recognized activity in the 3D space. Our system achieved around 20cm average localization error (in a 4 × 5 room) which is comparable to Kinect's body skeleton tracking error (10-20cm), but our system tracks activities instead of Kinect's location of people.
我们提出了一种深度学习框架，用于在动态且拥挤的现实世界环境中进行快速三维活动定位与跟踪。我们的训练方法与传统的活动定位方法相反，传统方法是先估计活动的可能位置，然后预测其发生情况。相反，我们首先使用深度视频和射频识别数据作为输入，训练一个深度卷积神经网络用于活动识别，然后利用该网络的激活映射在三维空间中定位已识别的活动。我们的系统实现了约20厘米的平均定位误差（在一个4×5米的房间内），这与Kinect的人体骨骼跟踪误差（10 - 20厘米）相当，但我们的系统跟踪的是活动，而不是Kinect所跟踪的人的位置。
