TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
token-verse.github.io
TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
1–10 of 24 posts
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#2Feels like a moodboarding multiplier for some design disciplines, if these aren't cherry-picked / transfer to other domains.
Pretty interesting.
Seems like you could apply similar ideas to text too.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#3This looks like an excellent step towards being able to apply consistency to generated images across a series.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#4If it worked with people, they’d show people.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#5If it worked with people, they’d show people.
The second example in the "Results" section includes a human.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#6Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#7Earlier quoted context omitted.
The second example in the "Results" section includes a human.
They do not show a realistic photo face transfer. They blur out the faces even. It would be a huge invention, but they did not achieve that.
Below the first Results header is a carousel of images. If you tap the arrows you can explore — I believe there are three examples where the final image is a person who’s face was applied from a reference photo.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#8Earlier quoted context omitted.
They do not show a realistic photo face transfer. They blur out the faces even. It would be a huge invention, but they did not achieve that.
Below the first Results header is a carousel of images. If you tap the arrows you can explore — I believe there are three examples where the final image is a person who’s face was applied from a reference photo.
Yes. But. The reference photo is blurred. The smallest details matter for faces! That's the whole point. I have no doubt you can do a kind-of-looks-like faces. But this is the same issue since Dreambooth. All the IP transfer approaches, even the best like Ideogram's, are failing on faces.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#9This looks like an excellent step towards being able to apply consistency to generated images across a series.
It looks as if it would trivially integrate into Whisk, which already has a similar feature for defining an outputs “subjects” “scene” and “style”.
Re: TokenVerse: Multi-Concept Personalization in Token Modulation Space by Google
#10Earlier quoted context omitted.
Below the first Results header is a carousel of images. If you tap the arrows you can explore — I believe there are three examples where the final image is a person who’s face was applied from a reference photo.
Yes. But. The reference photo is blurred. The smallest details matter for faces! That's the whole point. I have no doubt you can do a kind-of-looks-like faces. But this is the same issue since Dreambooth. All the IP transfer approaches, even the best like Ideogram's, are failing on faces.
There's two images where the face is transferred to the final image. The references images with blurred faces are all being used for a different reference; the pose, or "necklace", etc.
The faces are blurred in every image unless they explicitly want the face transferred to the final image, at least that's how it seems.