Convolutional-Block-Attention-Module icon indicating copy to clipboard operation
Convolutional-Block-Attention-Module copied to clipboard

Published 20 hours ago • verified publisher icon Youngkl0726

Can you provide the resnet50+CBAM code with the pre-trained weight?

Open liudaizong opened this issue 6 years ago • 3 comments

I want to compare the results with the SEblock, thx!

liudaizong avatar Sep 05 '18 07:09 liudaizong

I can provide the pretrained weight, but the model I trained did not reach the accuracy mentioned in the paper, even worse than seresnet50.

Youngkl0726 avatar Sep 07 '18 03:09 Youngkl0726

Actually, i just want to use CBAM+resnet50 or SEblock+resnet50 for baseline. Do you have the seresnet50 code with pretrained weight on ImageNet in pytorch? The performance of seresnet50 in https://github.com/Cadene/pretrained-models.pytorch is not good enough.

liudaizong avatar Sep 07 '18 05:09 liudaizong

How can I add the CBAM into YOLOv4. Can anyone suggest me the changes that I need to do in YOLOv4 code ? Or anyone have pretrained weights for the YOLOv4 with CBAM ? Thank you !!

Rvv1296 avatar Jun 22 '22 05:06 Rvv1296