P: Upload a reference image(s) for text to image

It would be great to be able to apply styles and text to image based on an existing image like in the service Dream by WOMBO 

Merging your thread here so we can track the demand for certain features.

Hello Adobe Firefly team,

It could be nice if we could upload an image has reference to be more specific on what we want.

By tha way, it's really cool! 😉

M

Merging your thread here so we can track the demand for certain features.

add the ability to add your photos

Merging your thread here so we can track the demand for certain features.

I am writing to share a suggestion that could help improve the program. As a user, I believe that adding the option to use one or more reference images could greatly assist the AI in creating more accurate and appealing visuals. While the prompt text is useful, it has its limitations, and incorporating images could provide a more robust and nuanced input for the system to work with.

I understand that implementing such a feature may require additional resources and development time, but I think it would be a worthwhile investment in the long run. As someone who values the program and its capabilities, I wanted to share this idea with you in the hopes that it could be considered for future updates.

Thank you for your time and attention, and I look forward to hearing your thoughts on this matter.

Best regards,

Bruno Baptista

Merging your thread here so we can track the demand for certain features.

Con igual promt, en otras plataformas de texto a imagen, varía mucho el resultado si se agrega una imagen de referencia, aproximandose mucho mas al resultado que uno desea obtener con el promt. Gracias!!

Merging your thread here so we can track the demand for certain features.

Upload Funktion eigener Fotos die als Referenz genutzt werden können oder als Basis für neue Kreation dieser fungiert

Merging your thread here so we can track the demand for certain features.

Agreed with everyone here - Firefly absolutely needs to have img2img. Big miss to only have txt2img.

It would be great to input an image of a face and have an output of different versions of that face with various art styles. This would be helpful for taking C-Suite and employee photos and making them look very interesting for adding to banners and various graphics for our presentations and event apps. 

Hi, The Adobe Firefly shows the ability to access a toolbar to edit and also the capability to upload an image to incorporate into the AI image. I am not seeing either one of these options. Can you please resolve this? Thank you. Jill

I would like the option of using a photo instead of, or in addition to, a text prompt. This is a feature of several other AI image generators, and I’m sure you plan on adding it to your finished product.

My suggestion is to include a slider which determines how closely or loosely the generated images resemble the prompt image. At the minimum setting, the AI would produce realistic-looking results with relatively small variations from the original, changing only the color or size of some objects, and perhaps the lighting. At the other extreme, the AI would run wild and produce a variety of fantastic derivations, very freely related to the image prompt. These would include graphic elements not present in the original. Additional text prompts could be added, but wouldn’t be required.