The invention discloses a pull-up test system based on face recognition and human body posture estimation, and aims to solve the problems of accurate counting and cheating prevention in a pull-up test. The system is composed of a camera, a camera supporting column, a pull-up checker and a pull-up checker supporting column. The pull-up checker is composed of a mainboard case, a display screen, a main switch, a power supply assembly and a mainboard. The mainboard case accommodates the mainboard, the power supply assembly, and the display screen and the main switch are installed; pull-up softwarecomposed of an acquisition module, a newly-built input module, an identity comparison module, a counting module, a display module and a database is installed on the mainboard. The newly-built input module and the identity comparison module carries out face recognition through a face recognition algorithm, the counting module estimates the posture of a human body through a human body posture estimation algorithm, and pull-up counting is carried out through video frame pixel coordinates of the double wrists, the nose and the neck of a tested person. The system is reasonable in design, and can accurately detect the number of the pull-up primers and prevent test replacement.