Height is an important human parameter used in many fields such as healthcare, beauty, sports, etc. There are many different non-contact height measurement methods in the world, but most of these methods only support measuring a person’s height in a standing straight posture, at a fixed distance. Therefore, in this study, we developed a system to support measuring human height in a bent-knee posture at different distances by extracting the coordinates of human skeletal landmarks using MediaPipe Pose Estimation. Moreover, our system also uses a multiple linear regression model to improve the accuracy of human height estimation. From the experimental results, our model has a relatively low error of about 2.46 cm (~1.44%). In the future, the research will be extended to support measuring human height in different postures such as sitting, walking, lying down, etc.