100,000 Faces - Free resource of 100k diverse faces generated by AI

We are happy to share with you a massive free resource of 100k faces, generated from scratch using AI.
This project showcases a glimpse into the technology we are building to create media on-demand.

Add a comment

Replies

Best
Hello! Do you have a link to your exact licensing terms for generated images?
I see. Thanks. That makes the images completely useless for pretty much anything, but at least now I know.
Cool project, I guess AI has a hard time aligning the eyes though?
it has! Maybe because I'm cross-eyed myself, haha. The cure is segmentation + more training data.
StyleGAN is fun. I was playing around with it a bit. For example here the images were created using StyleGAN from RunwayML’s desktop client on their “remote GPU” (your repository could have saved me that step :):) ). SqirlZ-Morph was used for the face morph transitions. At the end there is a manual mix of 3 different style transfer images. The facial animation was made with MugLife. Video was edited with Videoleap, Audio was created with SampleBot.
Judging by the comments of an AI modeling agency sounds awfully close to - looks like you made some of the same improvements on the StyleGan we have been working on, really cool!
Maybe the public will finally realize what AI can do ...
So who is going to the models that they don't need to come to work on Monday?
Is there anything hidden in the image that identifies it as "generated"? If no please consider.
how about this: we need a technology that certifies the image remains unchanged. Crime scene photography, product shots, passport photos — we need a technology that confirms the final shot is the result of the light passing the lens, without any manipulation.
for the rest, let's assume it's photoshopped, beauty-filtered and otherwise partially or entirely generated.
The adults look great, you need a LOT more diversity, and the children look...too old. Something about the eyes and nose? A lot of research shows that mammals have a lot of neurons that recognize how old young mammals are, both their own and other species.
I agree. We need more training data. For that, we need more children in our studio. Children are harder to deal and get tired quickly. But we need them. Challenge accepted.
Having to credit the site when using one of their AI images is a dealbreaker for commercial use. I can't see any client wanting ugly credit links displayed on their print designs, product designs, websites, etc. It's just not professional. Imagine including AI generated portraits on your client's "about us" web page, only to have visitors / customers see that the portraits are fake AI generated people when they click on the required credit link. I think your technology is impressive, but your business model makes no sense to me. Either offer an option to purchase images at a fair price, or provide them with no strings attached. Requiring folks to clutter their pages or products with credit links back to your website is a dealbreaker for me, and for most users I assume who are interested in the technology for business related purposes. I'd seriously reconsider your approach.
makes sense. Will do once we sort some stuff.
Hola feliz en poder ver nuestra tecnología