search

liuzhengzhe / DreamStone-ISS

This is the project for DreamStone: TPAMI & ISS: ICLR 2023 spotlight
https://liuzhengzhe.github.io/DreamStone.github.io/
43 stars 3 forks source link

Question about quantitative comparisons #9

Closed WayneCV closed 9 months ago

WayneCV commented 9 months ago

Hi, when computing the FID, you randomly sample 2600 images in the ShapeNet dataset as GT. So when you compute the FID about CLIP-Forge and DreamFields, did you also use the same GT (from ShapeNet)? Thanks in advance.

liuzhengzhe commented 9 months ago

Yes.

Zhengzhe

WayneCV @.***> 于2023年12月1日周五 16:56写道：

Hi, when computing the FID, you randomly sample 2600 images in the ShapeNet dataset as GT. So when you compute the FID about CLIP-Forge and DreamFields, did you also use the same GT (from ShapeNet)? Thanks in advance.

— Reply to this email directly, view it on GitHub https://github.com/liuzhengzhe/DreamStone-ISS/issues/9, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB7WPDCHVSG4LT74R763CKDYHGLTXAVCNFSM6AAAAABACP362SVHI2DSMVQWIX3LMV43ASLTON2WKOZSGAZDANBRGQ2DGNY . You are receiving this because you are subscribed to this thread.Message ID: @.***>

WayneCV commented 9 months ago

Yes. Zhengzhe WayneCV @.> 于2023年12月1日周五 16:56写道： … Hi, when computing the FID, you randomly sample 2600 images in the ShapeNet dataset as GT. So when you compute the FID about CLIP-Forge and DreamFields, did you also use the same GT (from ShapeNet)? Thanks in advance. — Reply to this email directly, view it on GitHub <#9>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB7WPDCHVSG4LT74R763CKDYHGLTXAVCNFSM6AAAAABACP362SVHI2DSMVQWIX3LMV43ASLTON2WKOZSGAZDANBRGQ2DGNY . You are receiving this because you are subscribed to this thread.Message ID: @.>

Thanks!