search

wl-zhao / VPD

[ICCV 2023] VPD is a framework that leverages the high-level and low-level knowledge of a pre-trained text-to-image diffusion model to downstream visual perception tasks.
https://vpd.ivg-research.xyz
MIT License
509 stars 31 forks source link

Download pre-trained checkpoints #64

Open caoer01 opened 2 weeks ago

caoer01 commented 2 weeks ago

I wanted to download pre-trained checkpoints, but the URL（ https://github.com/benjamin-paine/stable-diffusion-v1-5） was no longer valid, so I found the new URL(https://huggingface.co/benjamin-paine/stable-diffusion-v1-5/tree/main), which one should I download?Or is there any kind and handsome friend who has saved it and can provide network disk resources? ![Uploading 截图 2024-11-04 10-51-14.png…]()

caoer01 commented 1 week ago

checkpoints/v1-5-pruned-emaonly.ckpt