clarification AMOTA

[dmatos2012]

Hello,

A brief question on AMOTA definition: Per the paper and the code, AMOTA is :

So, what happens, if my model only has true positives with confidence values lets say from 1 to 0.50 instead of 1 to 0. In the current code, summation of MOTA is always divided by L=40 as per the paper.

However, if I only have 24 recall values (recall = 0.60) is the max, is it still correct to divide by 40 or should you divide by 24 then?Asking as currently, if i have different confidence scores, I get different AMOTA due to having less recall values, but always dividing by 40, and not sure whether thats intended behavior. I would think since its L = length of recall values it should be then divided by 24 in this case, rather than 40 regardless of recall values.

Would love clarification.

Thanks

[xinshuoweng]

if you can only reach recall = 0.6, I.e., not detecting the other 40% objects, you still have MOTA values at all 40 recall values, which are just 0s at the other 16 recall values. So the overall AMOTA score will be low because it is averaged with many 0s, which means that the metric is penalizing systems with a low recall.

Instead, if you divide by 24 recall values (values that are larger than 0, ignore the other 16 0s), you are not penalizing an MOT system with a lower recall.