-
Notifications
You must be signed in to change notification settings - Fork 202
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
improve text-to-image task page #889
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks a lot 💗💗
|
||
### Personalization | ||
|
||
Personalization refers to techniques used to customize text-to-image models, where we introduce new subjects/concepts to the model so that we can then use the model to generate new images of those subjects with a text prompt. For example, one can use these techniques to generate images of themselves, using as little as one reference image. These include teaching the model a new concept both in training free manner or through fine-tuning. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Personalization refers to techniques used to customize text-to-image models, where we introduce new subjects/concepts to the model so that we can then use the model to generate new images of those subjects with a text prompt. For example, one can use these techniques to generate images of themselves, using as little as one reference image. These include teaching the model a new concept both in training free manner or through fine-tuning. | |
Personalization refers to techniques used to customize text-to-image models. In this technique, we introduce new subjects/concepts to the model and then use the model to generate new images of those subjects with a text prompt. | |
For example, one can use these techniques to generate images of themselves, using as little as one reference image. These include teaching the model a new concept both in a training-free manner or through fine-tuning. |
I am actually irritated by personal DreamBooths, to be honest, especially with the recent Korean cases where it's weaponized against women. Can you change the example?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That's a good point. But I also feel like it's such a major use case, it's a bit odd not to mention it. Cause mentioning style Loras can also potentially be problematic on the artists angle
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
perhaps mentioning both briefly?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These are good points. A couple of alternatives:
- Use "your dog" as an example (I did this in my suggestion). I agree with Linoy that this somewhat sweeps a major use case under the rug.
- Be open about it and include a comment about using these features responsibly, respectfully, and ethically.
Co-authored-by: Merve Noyan <[email protected]>
Co-authored-by: Merve Noyan <[email protected]>
Co-authored-by: Pedro Cuenca <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you!
some improvements & updates to task page