Update crabs/tracker/README.md

Co-authored-by: nikk-nikaznan <[email protected]>
Signed-off-by: sfmig <[email protected]>

crabs/tracker/README.md

@@ -23,7 +23,9 @@ To compute the total number of false negatives (or missed detections, `FN`) at a
 
 To compute the total number of false positives (`FP`) at a given frame `f`, we count the number of detections that do not match with any of the ground-truth object defined at frame `f`.
 
-A true positive (`TP`) is defined as a detection that sufficiently overlaps with a ground-truth box (with overlap measured with the `IOU` metric). To compute the number of identity switches (`IDs`) at a given frame `f` we inspect the set of true positives, and check if for each of their ground-truth IDs, the predicted ID at frame `f` matches the predicted ID at the last frame the object was detected. If the predicted IDs do not match for the same ground-truth ID, we count that as one identity switch.
+A true positive (`TP`) is defined as a detection that sufficiently overlaps with a ground-truth box (with overlap measured with the `IOU` metric). 
+
+To compute the number of identity switches (`IDs`) at a given frame `f` we inspect the set of true positives, and check if for each of their ground-truth IDs, the predicted ID at frame `f` matches the predicted ID at the last frame `f-1` the object was detected. If the predicted IDs do not match for the same ground-truth ID, we count that as one identity switch.
 
 This is slightly different to some MOTA definitions, which only account for identity switches between consecutive frames. It is also different from other implementations, which define an "expected" predicted ID for each ground-truth ID. This "expected" predicted ID is the predicted ID that is most often (in terms of number of frames) associated to a ground-truth ID.