Giters

microsoft / RegionCLIP

[CVPR 2022] Official code for "RegionCLIP: Region-based Language-Image Pretraining"

Repository from Github https://github.commicrosoft/RegionCLIPRepository from Github https://github.commicrosoft/RegionCLIP

How much GPU memory do we need to run RegionCLIP

Xuefei98 opened this issue · comments

commented

First of all, great work and thank you for open source all the code! I wonder how much gpu memory do we need to run RegionCLIP?
Thanks,

Xuefei

commented

It depends on how large the batch size is. Since our model adopts the architecture of Faster RCNN, it requires a similar GPU memory as Faster RCNN. You can refer to the model zoo in Detectron2 for more details.