๐๏ธ Cityscapes Semantic Segmentation
Model:
SegFormer-B2 ยท
Training:
80 epochs ยท
Backbone LR:
6e-5 ยท
Loss:
0.7 CE + 0.3 Dice
Try it yourself
Dataset results
Try it yourself
Dataset results
Upload a street image
Drop Image Here
- or -
Click to Upload
Run โถ
Prediction mask
Examples
Colour legend โ 19 Cityscapes classes