search

uncbiag / SimpleClick

SimpleClick: Interactive Image Segmentation with Simple Vision Transformers (ICCV 2023)
MIT License
209 stars 32 forks source link

COCO and LVIS dataset preprocessing #10

Closed MaureenZOU closed 1 year ago

MaureenZOU commented 1 year ago

Hi Authors! Thanks so much for your great work : )

I have a question on how do you combine the COCO and LVIS dataset, do you replace the same mask in COCO with LVIS mask? Thanks a lot!

Best, Xueyan

qinliuliuqin commented 1 year ago

Hi Xueyan, please refer to Sec. 4.2 of the RITM paper.