search

Verg-Avesta / CounTR

CounTR: Transformer-based Generalised Visual Counting
https://verg-avesta.github.io/CounTR_Webpage/
MIT License
92 stars 9 forks source link

Count in Case of Multiple categories #15

Closed jaideep11061982 closed 1 year ago

jaideep11061982 commented 1 year ago

@Verg-Avesta so far your model working good on our images. I have this q 1) If suppose there are multiple categories of objects present in counting area say Cat A-4, Cat B-5 , Cat C-10 What is going to be count returned by the model, can it provide Category wise count as shown above or total count i.e 19 in this case.

2) For better counting results do I have to give atleast 1 BBX from each cat as assistance ?

Verg-Avesta commented 1 year ago

  1. The counting results depend on how different these categories are and what exemplars you provide. There are no deterministic results about what the model will count when the object categories are similar.

  2. You don't have to do this this because the model itself can generalize to a certain degree, but providing more representative exemplars will very likely to improve the performance.