SOTA Semantic Segmentation

I would like to know what are the State of The Art Architectures for Semantic Segmentation.

I know:

  • UNet
  • DeepLabV3+
  • Mask-RCNN
  • 100 Layers Tiramisu

Do you know other architectures?

¿EfficientDet? There are several Pytorch implementation like this.

Nice addition.

However, I would like to focus the discuss in high precissions ones