Add SwinV2(UperNet) and FD-SwinV2 to model_zoo

[rznas]

Describe the feature

Motivation There is no implementaiton of SwinV2 for semantic segmentation Related resources The original implementation is only for Image Classification.

Additional context

Swin Transformer V2: Scaling Up Capacity and Resolution.

Contrastive Learning Rivals Masked Image Modeling in Fine-tuning via Feature Distillation.

[MeowZheng]

We are planning these models indeed. However, it won't be finished in a short time due to the lack of developers. Collaborations are appreciated If you are interested in it.

[rznas]

That's fine, I'm on it 😊