Feature Requests

The goal is to develop an Adobe Firefly feature that allows real-time creative collaboration and feedback among several designers and stakeholders. With the help of this tool, designers from different locations might collaborate easily on the same design project.The notion of "Dynamic Design Collaboration" may have the following key components:Real-time editing enables collaborative design processes by allowing several designers to work on the same design file at once and view each other's changes in real time.Feedback & Commenting: Designers and other interested parties can write comments directly on the design canvas, offering pertinent commentary and kicking off debates about certain design features.Version Control: The functionality would come with an effective version control system that enables designers to trace changes, go back in time, and seamlessly combine design iterations.Design Presentations: Using Adobe Firefly, designers may build interactive design presentations th

I think a section would be good... about real beings... fantastic beings... objects... to better delimit the result..... as well as buttons for this....

Hi! I do alot of product rendering for industrial design and I think it would be super useful to be able to take two images, for example, a png of a product like a coffee machine, and an image of a kitchen, and be able to put the coffee machine into the context of the kitchen by blending the two images lighting, clarity ext.  Or even take a 3dm file and ask firefly to put it into the context of an image and match perspective ext.  Cheers!

I want to be able to upload a group of images to set the tone and select components of each that I want to highlight.Example1:I want the wheels from one image put on to the car in another image with the background of a thrid image and the syle or lighting of a fourth image. Include the person in another image and composit them together in realistic photo.Example2: Here is a group of pins (images) generate more images that fit that style.

I would recommend creating a more immediate mechanism for the user, which allows you to rework the newly generated image taking into account the "likes" placed on previous results. In order to improve and address more accurately the new generated images.

Hi being a minority artist I'd like to see more diversity in image generation...meaning body shapes, heights, race, disabilities etc.... just a suggestion though...I appreciate you listening...I'd like the definition of beauty to include many things and not just one mold:) I did see different models already...so great...but even more diversity would be awesome...a rainbow is more beautiful with more colors not just one:)

This has great potential! The preset fonts are good bases for this concept, but rather than loading more fonts in the future, it would be interesting to impliment vecotor Illustrator files or pdfs with transparent backgrounds in place of the black text.

I think it would be super amazing is if a feature is that one could use a generated image and/or text and firefly could turn it into an 3d object. 

Create a feature where you describe a story and the AI ​​creates comic-style images based on the text

I had this unique idea. Sometimes our selections for generative fill are not perfect, or the original image does not blend properly with the generated objects. It would be unique if say, I asked for a sunset to replace the sky in the image, and then it would warm up and alter the rest of the image to fit the 'theme' or color tones of the generative fill. That way the original image blends better with the generated aspect!

It would seem possible for an AI to analyze frames of a  damaged film and replace faded color or expand the frame using similar undamaged frame info to recolor or outpatient. This would be a game changer in the film and video restoration arena. 

Is there any chance to have a limited rotate left/right or up/down within Firefly image?  I believe it would be helpful to get a different persprctive view, before saving or adding next text prompt. Otherwise, I LOVE LOVE Firefly thus far. BT

in order to search for a specific style , we could add a input field Search funcitonlity that quickly filters thorugh the StylesNice to Havea typeahead fucntinality could be included.on focus, a list of the user's most used styles could appear as suggestion in a dropdownPros:this will make it easier for users to search for specific stylesif the list of styles grows in the future, it owuld be easier to find them

Create a text effect based on thunders and lightnings based on blue color. 

Hello,It would be nice to be able to rotate and flip horizontally and vertically the generated fills. I was trying to add a persons sihloette, but it was facing the wrong direction even when I typed the direction I wanted it facing. Being able to have controle of this would be awesome. Thank you for your time.

I would like a get idea about the text to image. It would be nice if the generator was most specific. With more options, more prompts and more didactic. It woult be really really good. Ass,Erick Gonçalves 

This is more of a new AI tool request than a feature request: I have an actual situation that I think AI could be used to solve. I am just starting to edit a feature film in Premiere Pro that has 40+ hours of footage. I will most likely take a month to manually search and ad meta data tags to the footage before I can even start the creative part of editing. It would be awesome if Adobe had an AI tool that couldhelp add meta data for me.  For example: It would be cool if I could “teach” the AI engine who each talent is: “Actor 1”, Actor 2”, “Actor 3”,…. Also tag: “day shot”, “interior”, “night”, “exterior”,…..Possibly teach the AI to identify locations too. I’m already a big Adobe fan. If you come out with a “Video Meta-Data AI Tagger” tool I will become an Adobe evangelist! Thanks,Rich Jackson

Just need a option to give Proportion of the subject and the support items either in percentage or ratio some thing like that

Imagine asking Firefly to generate a person. That person looks fine, but you'd really like it if they were looking in a certain direction, or at a particular object. By using this feature, the artist could mask a certain area (or the whole photo), input the rotational changes, and then preview generations. This could be a very complex feature, but the basic idea is this:Load an imageMask the element you want re-orientedInteract with a context menu with XYZ rotational, focal length, and maybe QoL presets/swatches to select default parameters (ie. side, top, back views).Generate the output with multiple variations (as it currently does) Bonus parameters to includeScale (ie. the difference between a pie on your plate and a pie the size of a city)Cohesion (ie. how much should the output look like the original reference)Directional Lighting

It would be really nice if it one day understood what '60" x 60" of surface area depicted at a resolution of 2000 x 2000 pixels @300 dpi' means.

red hair

I will come directly to my points: I had a bug where I could not 'insert' new objects in my imagesBackground removal is amazing. however, i would suggest giving the user the choice to choose which image to keep. If there are three person standing on the road, the user can choose only two person standing or the user can choose that only one person and the road stays and the rest disappearsWhen I first uploaded my image, it was small. I had to zoom in using my touchpad. Make it easy by adding the sliding bar Add a UI to undo and redo

The scene depicts a young woman engrossed in her work, seated at a computer amidst a breathtaking mountain landscape. A magnificent waterfall cascades down nearby, providing a soothing symphony of rushing water. Nestled amidst the picturesque scenery, a charming wooden house exudes a rustic charm. In this enchanting grove, a friendly dog accompanies the girl, adding an element of warmth and companionship. Every intricate detail of this magical setting is brought to life, captivating the imagination.

What is White tiger 

What is Whitetiger