> @yangyangsu29 after model convergence, how many the loss will be?

bradyz / cross_view_transformers

Cross-view Transformers for real-time Map-view Semantic Segmentation (CVPR 2022 Oral)

MIT License

525 stars 80 forks source link

> @yangyangsu29 after model convergence, how many the loss will be? #27

Closed hengguan closed 2 years ago

hengguan commented 2 years ago

@yangyangsu29 Did you solve the problem now? Thanks. training each class （vehicle and driveable area）separately，it can reproduce the paper‘s result indeed； in visualization, the predict‘s map overlay together as follow：

road_19

_Originally posted by @yangyangsu29 in https://github.com/bradyz/cross_view_transformers/issues/16#issuecomment-1157269989_