New heads and backbones for object detection 🤔

[Nexer8]

Hello, given the number of heads and backbones for the task of object detection remains rather low in comparison to MMDetection, I'd like to ask whether there are any plans to introduce more in the near future?

Also, are there any plans to introduce backbones using transformer architecture, such as Swin Transformer, into this project, or is the lightning-transformers meant particularly for this?