Tripping is considered a major cause of fall in older people. Therefore, foot clearance (i.e., height of the foot above ground during swing phase) could be a key factor to better understand the complex relationship between gait and falls. This paper presents a new method to estimate clearance using a foot-worn and wireless inertial sensor system. The method relies on the computation of foot orientation and trajectory from sensors signal data fusion, combined with the temporal detection of toe-off and heel-strike events. Based on a kinematic model that automatically estimates sensor position relative to the foot, heel and toe trajectories are estimated. 2-D and 3-D models are presented with different solving approaches, and validated against an optical motion capture system on 12 healthy adults performing short walking trials at self-selected, slow, and fast speed. Parameters corresponding to local minimum and maximum of heel and toe clearance were extracted and showed accuracy +/- precision of 4.1 +/- 2.3 cm for maximal heel clearance and 1.3 +/- 0.9 cm for minimal toe clearance compared to the reference. The system is lightweight, wireless, easy to wear and to use, and provide a new and useful tool for routine clinical assessment of gait outside a dedicated laboratory.