Hsu Feng-Shuo, Su Zi-Jun, Kao Yamin, Tsai Sen-Wei, Lin Ying-Chao, Tu Po-Hsun, Gong Cihun-Siyong Alex, Chen Chien-Chang
Bio-Microsystems Integration Laboratory, Department of Biomedical Sciences and Engineering, National Central University, Taoyuan 320317, Taiwan.
Department of Psychiatry, Taichung Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Taichung 427213, Taiwan.
Entropy (Basel). 2023 Feb 11;25(2):336. doi: 10.3390/e25020336.
Fusing object detection techniques and stochastic variational inference, we proposed a new scheme for lightweight neural network models, which could simultaneously reduce model sizes and raise the inference speed. This technique was then applied in fast human posture identification. The integer-arithmetic-only algorithm and the feature pyramid network were adopted to reduce the computational complexity in training and to capture features of small objects, respectively. Features of sequential human motion frames (i.e., the centroid coordinates of bounding boxes) were extracted by the self-attention mechanism. With the techniques of Bayesian neural network and stochastic variational inference, human postures could be promptly classified by fast resolving of the Gaussian mixture model for human posture classification. The model took instant centroid features as inputs and indicated possible human postures in the probabilistic maps. Our model had better overall performance than the baseline model ResNet in mean average precision (32.5 vs. 34.6), inference speed (27 vs. 48 milliseconds), and model size (46.2 vs. 227.8 MB). The model could also alert a suspected human falling event about 0.66 s in advance.
我们融合目标检测技术和随机变分推断，提出了一种用于轻量级神经网络模型的新方案，该方案可以同时减小模型大小并提高推理速度。然后将该技术应用于快速人体姿态识别。分别采用仅整数运算的算法和特征金字塔网络来降低训练中的计算复杂度并捕捉小物体的特征。通过自注意力机制提取连续人体运动帧的特征（即边界框的质心坐标）。借助贝叶斯神经网络和随机变分推断技术，通过快速求解用于人体姿态分类的高斯混合模型，可以迅速对人体姿态进行分类。该模型将即时质心特征作为输入，并在概率图中指示可能的人体姿态。在平均精度均值（32.5对34.6）、推理速度（27对48毫秒）和模型大小（46.2对227.8MB）方面，我们的模型比基线模型ResNet具有更好的整体性能。该模型还可以提前约0.66秒发出疑似人体跌倒事件的警报。
