You patch it the same way you patch anything in Max.
1. What range of data values (float/int) does the granularized patch expect?
2. We’ve got no idea what you’re using for color tracking, but you need to find out the range of data values the color tracking stuff spits out for whatever kind of tracking you’re doing.
3. Compare the ranges. Do you need a scale operator to convert the output data range from whatever color tracking you’re doing to the expected input data ranges? If so, put one in for the two granularized patch parameters you want to run.
4. Listen and rethink.