Feature Requests

Make it better than midjourney and Leonardo.ai check the images these two image generators can out put. Your ai generates disgusting images. I am sad to say that truth hurts. 

It would be nice to be able to combine two or more images together by describing which part to keep on each image.The main interest would be to be able to upload your own image as a base for a generated image.

It would be interesting to be able to ask to describe a character or a scene from a specific book (so that the AI ​​keeps in mind that it is that character from that book) or a historical character based on an already known historical description

Would be great to know more about the optimal syntax. e.g. how to insert letters and numbers for a lettering or logo with keeping the exact phrase.

Firefly as an AI editor on mobile platforms like Android or IOS. Basic video and image editing, as a "plugin" and update by inserting the feature into existing applications such as Adobe Rush and LightRoom (not classic).Sample: "low saturation, soft green light, overlay with yellow light coming from the left side of the screen"

For example when I type "Griffin flying in the sky.", it gives me a regular eagle without the lion body.I'd love to do more monster/creature images, and not just dragons.Please add this, so we can be more creative with our images. ❤️If this wasn't already planned, that is.

In the furniture world it is extremely hard to get images of specific pieces of furniture covered with all the coverings of fabric and leather available. It would be awesome if ai could take our photo of a piece of furniture and recover it with our photo sample of available fabrics and leathers

Adding an existing image for reference on the start might open more realistic images.

In the "Edit -> Insert" mode, please add the option to insert a reference image as well, not only a text prompt —so that we can specify: "Insert this specific image at this specific location."

Faceswap feature like insightfaceswap that you can use on midjourney, it's extremely accurate.

Hi everyone!I love Firefly beta but for me it would be nice, if I could load up videos or own pictures/photos and change them with firefly and then afterwards download it.Is that possible? I hope so! Thanks ❤️

Upload subject or scene images, for the AI to reference in a text prompt. Example:Uploaded Image Marked subject (shows a guy in a blue shirt with a pattern and grey kakis waving hi) Uploaded Image Marked scene (shows a beach scene)Text prompt (Show the "subject" walking in the "scene" with an ice cream cone)The visual provided would reference the images uploaded for a more direct result.

Bonjour,pourquoi ne pas rajouter la possibilité a l'ai de conceptualiser des logos, et autres illustration graphiques comme des etiquettes ?

Not able to genrate image of particular company products or showroom or place, say for example u tell it change background like sitting in a indgo plane or standing at a apple showroom it fails to understand the background 

After removing the background, there is no option to export the image with the transparent background

Seria incrivel criar uma imagem com uma imagem de referencia que possamos fazer upload, seja uma imagem ou uma foto como por exemplo um desenho.dd

This AI is an amazing tool, thanks for making it!Maybe add an option to directly export it to Ps or Ai?

I think it would both be consumer friendly and an attractive service for people to pay a set price(not subscription which is anti-consumer) to remove watermarks at a price or at least a set number of images fora set price.  To say that Artiificial Intelligence has improved within the past few years would be an understatement but much like any other market, there's potential to have a quick cash grab.  I think it would be in the interest of Adobe to offer services that are not only competitive in the AI image generating market but that also stand out.  With that, I'd also like to suggest upscaling, downscaling, smoothing reduction and choices of resolution as features as well.   

Is there a way to generate an image with a transparent background. Altrnantively, will there be a way to produce an image as file with layers so we can easily  remove the background? 

Ai to have knowledge of pop culture, for an example it should know about spiderman. or what disney or marvel is.

The AI could take people from 3 different photos and connect into one.

We are also having some questions on superscript and subscript. We are trying to implment the superscript and subscript for the dynamic text in a text area. We have tried even with antialaising and still the super script is not visible properly. Can we have some source code on how we can implement this.

Auto select an object,,like sky, dresses, person, sea, tree etc.

I am kindly request you to please add famous celebrity faces places just like Robert junior Ven Diesel Shahrukh khan Salman Khan ...    This will be help to people to take a photos With their favorite celebrities   Thanks    ["places" changed to "faces" as per later post from OP]    

Bonjour, bravo pour ce que vous faites.  J'aimerais bien que un jour on puisse obtenir les api de Adobe Firefly car J'aimerais développer un bot télégram pour mes différents groupes et chaînes télégram.  Merci