ITER estimates they need 250 kilos of hydrogen per year for a 100MW plant, half deuterium half tritium. That's 125k/2 + 125k/3 = 103k moles of gas, so ~100k * N atoms of hydrogen. The interstellar medium density is given by the wiki as 10¹² molecules/m³, of which 70% is hydrogen. So, we need to absorb 100k moles * N / 10¹² = 6*10^16 cubic meters of ISM to get enough hydrogen for a 100MW plant on board the ship for one year.
And this assumes that the hydrogen is available in exactly 50/50 deuterium/tritium combinations, which seems extremely improbable given that tritium has a very short half life (a few years at most).
I very much doubt that refueling from the ISM is plausible.*
100MW is more than the energy consumption of 50,000 homes. To collect 6*10^16 cubic meters/year at an average speed of 100km/h (walking pace in outer space), you'd need a "net" of less than 10km x 10km, which is completely plausible for a ship that large.
If you have a small ship using the reactor for thrust, at 1% of the speed of light the width of the "net" is much smaller.
And tritium is used in existing designs but we're still in the infancy of the technology. Stars fuse deuterium with neutron-free hydrogen (protons), and make the deuterium themselves.
