You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
when I use it to train my own model, it cannot work well. whatever the precision it is, its lowest recall rate is unexpectable. Could you share your training experience? Thanks a lot.
And the another question is that this method can be adopted at vehicle detection?
The text was updated successfully, but these errors were encountered:
@kingloo2014 hi, recently, i have read an issue about vehicle detection whose method is NPD+DQT, I try to train my own model, but the result is also unexpectable. have you trained out a good model about face or vehicle? Could you share some experiences? Thank you for your share. @cmxnono this is my trainning parameters: NPD_Project.exe -faceDB pos_car.list -negDB neg_car.list -outModel result_car -objSize 24 -numPos 6162 -negRatio 1 -maxTreeLevel 8 -minDR 1 -maxFAR 1e-15 -maxNumStages 200. Does it has some mistakes? Please help me and thank you for your guidances.
And please pardon my poor English.
when I use it to train my own model, it cannot work well. whatever the precision it is, its lowest recall rate is unexpectable. Could you share your training experience? Thanks a lot.
And the another question is that this method can be adopted at vehicle detection?
The text was updated successfully, but these errors were encountered: