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
To my understanding, two-stage detector in general is more accurate than single-stage detector. However for all the pre-trained models listed in ModelZoo, it has lower or similar AP score compared to models in Yolov5. I wonder what contribute to the difference since the pertained models in Detectron 2 are in general more complex and larger. Can anyone explain about this? Thanks!
For example, a big R101-FPN Faster-RCNN model only achieve AP 43, while yolov5x has AP 50.7, which is much smaller and faster.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
To my understanding, two-stage detector in general is more accurate than single-stage detector. However for all the pre-trained models listed in ModelZoo, it has lower or similar AP score compared to models in Yolov5. I wonder what contribute to the difference since the pertained models in Detectron 2 are in general more complex and larger. Can anyone explain about this? Thanks!
For example, a big R101-FPN Faster-RCNN model only achieve AP 43, while yolov5x has AP 50.7, which is much smaller and faster.
Beta Was this translation helpful? Give feedback.
All reactions