Kemal Oksuz
Kemal Oksuz
I agree it is not difficult to implement the network. But I think it could be better to have an implementation from the authors to be sure about all implementation...
Do you have mmdet v2 implementation of solov2 as well?
Many thanks! I will check it out.
> > @kemaloksuz An implementation of SOLOv2 with mmdet 2.3 can be found [here](https://github.com/atlantis-ar/mmdetection). > > We extended the SOLO branch from [here](https://github.com/WXinlong/mmdetection/tree/SOLO?rgh-link-date=2021-06-08T10%3A40%3A39Z). > > Thanks for sharing this, out...
Center Net predicts (i) detection confidence scores (i.e. heatmaps) corresponding for two corners and centers, (ii) embeddings to "group" predicted individual points as boxes and (iii) offsets for better localisation...
Hi! The reason why we adopted RetinaNet is our baseline, AP Loss, was trained on RetinaNet, and we did not want to replace it for fair comparison. We only replaced...
Are you able to solve this error?