Athlete's pose acquisition and analysis is promising to provide coaches with details of athletes performance and thus help to improve athletes' performances with more detailed supervision from coaches. Compared with traditional ways of acquiring an athlete's gesture, such as using wearable sensors, computer vision technology has advantages of low-cost, high-efficient and non-intrusive. This paper aims to bridge these two fields, by reconstructing athletes' trajectory using monocular (i.e. single-camera-shot) videos. Under a few assumptions that are applicable to most of the sports of athletics, we proposed a method combining computer vision techniques and physics laws to reconstruct athletes' trajectories from monocular videos. The method first estimates 3D pose of athletes from video inputs, then performs kinematic analysis on estimated poses to reconstruct the trajectories of athletes. We tested this algorithm on videos from the triple jump finals of the 2016 Olympics in Rio de Janeiro. We achieved a best performance with 9.1% mean average error when using ground-truth foot-ground contact signal and 21.4% mean average error when using predicted foot-ground contact signal.