Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

anyone try FPN with yolo #1611

Open
chenguanfu511 opened this issue May 11, 2019 · 4 comments
Open

anyone try FPN with yolo #1611

chenguanfu511 opened this issue May 11, 2019 · 4 comments

Comments

@chenguanfu511
Copy link

I see PAN for object detection, and the paper use FPN+PAN to augment the information flow, resulting greater score on the coco benchmark. So I wonder how to use the FPN+PAN, letting yolo have the same potential.

Path Aggregation Network for Instance Segmentation architecture

image

@AlexeyAB
Copy link
Collaborator

@chenguanfu511 Can you give URL to this article?

@MasterSuoCN
Copy link

@chenguanfu511 Can you give URL to this article?

https://arxiv.org/abs/1803.01534

@AlexeyAB
Copy link
Collaborator

@chenguanfu511 @MasterSuoCN

You can try to train one of these models: AlexeyAB#3114 (comment)

For example, train this model: https://github.com/AlexeyAB/darknet/files/3199657/yolo_v3_spp_pan.cfg.txt
by using this pre-trained weights: http://pjreddie.com/media/files/darknet53.conv.74

@pinnintipraneethkumar
Copy link

So, This cfg files contains both fpn+cfg , i would like to use only fpn is there any cfg file for that.

@chenguanfu511 @MasterSuoCN

You can try to train one of these models: AlexeyAB#3114 (comment)

For example, train this model: https://github.com/AlexeyAB/darknet/files/3199657/yolo_v3_spp_pan.cfg.txt by using this pre-trained weights: http://pjreddie.com/media/files/darknet53.conv.74

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants