About the pretrained backbone.

Open jasonseu opened this issue 3 years ago • 0 comments

Hi, thanks for your great work. The performance of ResNet101 based ML-Decoder on the MS-COCO is impressive. The training details in your paper show that the ResNet101 is pretrained on Open Images. Did you conduct experiment about the ResNet101 pretraining on ImageNet ?

Jul 13 '22 03:07 jasonseu