Fine sediment infiltration into a river bed is a physical process affected by different human actions and has several environmental, socioeconomic, and river morphology consequences. A theoretical model is proposed herein aiming to reproduce the fine sediment content depth profile resulting from the infiltration of fine sediment into an initially clean gravel bed. The model is based on the probability of infiltrating particles to be trapped in a pore throat formed by three bed particles. The model is tested against previous experimental results and is found to reproduce adequately the occurrence of the two infiltration mechanisms reported by previous studies: bridging and unimpeded static percolation. Theoretical depth profiles are found to underestimate fine sediment content at the bed subsurface (below 2–3 gravel diameter depth) compared to the laboratory results. This may be due to hyporheic flow that is not taken into account in our model. In flow experiments, the particles previously infiltrated and deposited might be destabilized by pore water flow and their fall down to the bed might be magnified.