针对现有步态周期检测方法易受拍摄视角变化影响的问题，提出了一种拟合下肢几何特征的多视角步态周期检测方法。首先，利用MediaPipe模型提取步态视频序列中的人体姿态拓扑图，简化了图像预处理过程。然后，通过分析行走状态下人体下肢姿态拓扑图中各关节点间存在的周期性动态变化规律，将左小腿与水平地面构成的倾角以及中髋点（mid-hip）到左、右脚踝的欧氏距离比值作为特征进行提取。最后，采用傅里叶变换将特征数据拟合为正弦函数，并基于拟合结果进行步态周期检测。相比于当前主流的步态周期检测方法，本文方法在正、背面视角以及斜视角下都取得了较好的检测结果。

A multi view gait cycle detection method fitting the geometric features of lower limbs is proposed to address the issue of existing gait cycle detection methods being susceptible to changes in shooting angles. Firstly， the human posture topology in the gait video sequence was extracted by the MediaPipe model， simplifying the image preprocessing process. Then， by analyzing the periodic dynamic change law between the joint points in the human posture topology map under walking state， the inclination formed by the left shin and the horizontal ground and the Euclidean distance ratio from the midpoint of the left and right hip joints to the left and right ankle are extracted as features. Finally， the feature data were fitted into sinusoidal function waves by Fourier transform， and the gait period is detected based on the fitting results. Compared with the current mainstream gait cycle detection methods， the proposed method has achieved good front and back view and strabismus angle detection results.