search

cvlab-epfl / erasing-road-obstacles

Detecting Road Obstacles by Erasing Them
8 stars 1 forks source link

consistency networks #3

Open JAYCHOU2020 opened 3 months ago

JAYCHOU2020 commented 3 months ago

Is there a GT label for the training of consistency networks in this paper？

adynathos commented 3 months ago

The training dataset for the discrepancy network is created by injecting objects onto the road. Therefore the GT is generated with the dataset as it is just the pixel mask of the injected object. Furthermore non-food areas are marked as out-of-ROI.

On Sat, Aug 10, 2024, 23:02 ImJustMe @.***> wrote:

Is there a GT label for the training of consistency networks in this paper？

— Reply to this email directly, view it on GitHub https://github.com/cvlab-epfl/erasing-road-obstacles/issues/3, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB5BIAATX5NTIQVPDDG64KDZQ3H37AVCNFSM6AAAAABMKLG5MSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGQ2TSMZZGM2DMNQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>