search

wangxiang1230 / OadTR

Code for our ICCV 2021 Paper "OadTR: Online Action Detection with Transformers".
MIT License
87 stars 12 forks source link

HOw to resize the features (Dataset) #26

Open pgupta119 opened 1 year ago

pgupta119 commented 1 year ago

hi @wangxiang1230 , As per the paper you have used different dimensions of input. I would like to know what did you get it ?

thanks

wangxiang1230 commented 1 year ago

different dimensions

Do you mean "query dim"? This refers to the dim of query token of transformer decoder.