You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for your impressive work. I have a small question and hope to get an answer from you. In your paper, you mention that the design of controlnext brings a significant improvement in both the parameter count and inference time compared to controlnet. Is there any objective metric to compare the performance of the two? In particular, I noticed that there isn't an official implementation of SVD for controlnet. Did you retrain the SVD version of controlnet in your work, or did you use tools like ComfyUI?
Looking forward to your response, and once again, thank you for your creative work!
The text was updated successfully, but these errors were encountered:
Thank you for your impressive work. I have a small question and hope to get an answer from you. In your paper, you mention that the design of controlnext brings a significant improvement in both the parameter count and inference time compared to controlnet. Is there any objective metric to compare the performance of the two? In particular, I noticed that there isn't an official implementation of SVD for controlnet. Did you retrain the SVD version of controlnet in your work, or did you use tools like ComfyUI?
Looking forward to your response, and once again, thank you for your creative work!
The text was updated successfully, but these errors were encountered: