yrcong / RelTR

RelTR: Relation Transformer for Scene Graph Generation: https://arxiv.org/abs/2201.11460v2
230 stars 45 forks source link

Training on Visual Genome dataset #11

Closed dneirfi closed 1 year ago

dneirfi commented 1 year ago

Hi. Thanks for sharing your nice work!

Do you have any pre-trained weights on full Visual genome dataset? (not split one)

Thanks.

yrcong commented 1 year ago

Unfortunately we didn't train it on the original VG. I think it's a bit noisy (many ambiguous classes).

Instead, we trained on OpenImages

dneirfi commented 1 year ago

Thanks for replying!