I still don't think you make sense lmfao but np The problem is not with training material, the problem is SD is trained on 512×512 mostly, now SD2 on 786×786, those are low resos and hard to see hands properly It'll be solved once reso of training pics goes up
INNOVATION
-
Cerebras Systems CEO Interview: Raising the Bar in AI
By
–
AI Business' @wyliewong recently interviewed our CEO, noting that “Cerebras Systems is one of those rare companies in Silicon Valley that continues to raise the bar – not easy in a region full of innovative startups.” Read the interview here:
-
Why Dreambooth Struggles Rendering Hands Despite Training Data
By
–
You say there is plenty of photos of hands available so that's why Dreambooth can't render hands properly? I don't get it
-
SD Resolution Limits and Future Improvements for Image Quality
By
–
Also a lot is it due to 512×512 limit of SD1.5, with SD2 u can do 786×786 so it gets better, but we're all upscaling so real details we will get in a few months when we can do 1024×1024 and 2048×2048 etc after, it will exponentially get better
-
Two AI Image Generation Apps: Photorealistic vs Artistic Styles
By
–
Both mine and @dannypostma
's apps are the best 😛 Danny's better for special arty styles right now, I'm focusing more on photorealistic now but Danny might switch to that soon too Anyway they're both great! Maybe try Danny's app first: http://
profilepicture.ai -
Avatar AI Achieves Photorealistic Progress in Daily Updates
By
–
Daily http://
avatarai.me photorealistic progress pics 2 -
Avatar AI Achieves Photorealistic Progress in Daily Updates
By
–
Daily http://
avatarai.me photorealistic progress pics -
Ludwig v0.6 Simplifies ML Model Operationalization with Inference
By
–
Operationalizing state-of-the-art #ML #models just got easier! With @ludwig_ai v0.6, you get an end-to-end #inference pipeline out of the box. Check out the blog to learn more. Sample code and #data included. https://
pbase.ai/3Hrdstf #machinelearning #deeplearning #pytorch
