Objective To observe the obstacle of mandible to the cervical spine with the aid of dual fluoroscopic imaging system, explore the optimal perspective Methods of in vivo kinematics of cervical spine, and verify the feasibility of cervical in vivo kinematic researches. Methods A dual fluoroscopic imaging system composed of two C-arms placed in different angles was utilized. X ray images of the cervical spine for five healthy volunteers (4 male, 1 female) with standing, flexion extension, twisting and bending positions were obtained with the C-arms in the angle of 90°, 60°and 45°, respectively. The obstacles of mandible to the cervical spine in different positions were compared and the obstacle degree of mandible to the cervical spine was evaluated to screen the optimal perspective Methods. The perspective images under conventional head rotation and coaxial trunk rotation were collected during cervical twisting to compare the obstacle of mandible to cervical images. Results There were significant differences in mandibular obstacle degrees among the three angles measured in standing, flexion-extension, twisting and bending positions（P<0.05）. With two C-arms in 45°, the cervical spine was best imaged with the least obstacle. With two C-arms in 60°, significant differences could be found between the coaxial trunk rotation and the head rotation group. Conclusions For in vivo kinematics study of cervical spine, the obstacle of mandible to the cervical spine can be minimized with C-arms of dual fluoroscopic imaging system in 45°, and the coaxial trunk rotation can decrease the mandibular obstacle degrees as compared with the head rotation, which satisfy the requirement of 2D-3D image matching.