what kind of semantic segmentation model mentioned in the paper appendix? thx

OpenDriveLab / ST-P3

[ECCV 2022] ST-P3, an end-to-end vision-based autonomous driving framework via spatial-temporal feature learning.

Apache License 2.0

280 stars 34 forks source link

Closed emilyemliyM closed 1 year ago

ilnehc commented 1 year ago

Sorry for the unclear part. It is MaskFormer with a Swin backbone.

emilyemliyM commented 1 year ago

Sorry for the unclear part. It is MaskFormer with a Swin backbone.

Really thanks for the reply!

SO after the training in the Mapillary dataset, it just get the semantic output for the Nuscenes dataset in a test mode ? Is that right?

Regards.

ilnehc commented 1 year ago

Correct.