How to understand the resolution of pretrained dataset refered in snip paper?

mahyarnajibi / SNIPER

SNIPER / AutoFocus is an efficient multi-scale object detection training / inference algorithm

Other

2.69k stars 451 forks source link

How to understand the resolution of pretrained dataset refered in snip paper? #186

Open lyw615 opened 3 years ago

lyw615 commented 3 years ago

Thank you for your outstanding work ,but there are some problems to consult. The resolution of ImageNet dataset is 224224. When the weights of object detection network pretrained on coco dataset loaded, should the resolution of pretrained dataset set to 480640 ？