Feature Requests

to color the logic of a Sketch

You should be able to chan ge the aspect ratio of a picture in adobe firefly. It could even be auto generated based on the image inputed  

It would be nice to have access to a shift button to make using the brush tool easier.

Implment a way to help the AI know what you want. For example, I was trying to replace a chair I was sitting in with the iron throne, throne of swords, swords. It try but really had no idea.

A feature on generative fill where you can make the image bigger, and let the AI fill in the rest of the image, people are doing this all over the internet and it's not possible in the web version.

it would have been nice if the program could generate idea descriptions, this could be of great help in idea creation in the form of products and the like

may it will more interesting if add a crop feature, where we can increase or decrease our artboard/image

after an Image has been generated in Text to Image, an Icon can show up, enabling the user to edit the generated image in Generative fill. 

by enabling aspect ratio change user can choose another aspect or crop the image , empty space will be regenerated, 

Create a tool that, based on an image, provides all the characteristics to generate a prompt and allows you to edit it in order to obtain a similar but different image.

Would be seriously cool to have a hybridize/blend feature in Firefly that allows you to pick two (out of the 4) results it gives you and be able to sort of find a middle ground when you enjoy two results instead of just picking one. 🙂 Firefly is super neat, realizing so much artistic potential with it.! Keep up the good work

Bring a complete text of a book and the AI will match it from beginning to end including adding a little content by itself (such as all rights reserved, title of the book, etc.)will mark the content he produced.Get a list of automatic designs after you have created the table of contents, and in the creation of the content itself, it will recognize where the main chapters are according to the AI and not according to the styles. [same as index].Choosing the best parameters when you export to PDF according to your document is easy.Ideas to improve your existing design, and their execution at the click of a button.Identify manual footnotes made in Word, and automatically identify footnotes built into the text.Creating InDesign lines and frames by lettering.Adjusting the design of one page according to the design of the previous pages in your document, or according to your previous documents.Fill dummy text at your request like CHATGPT.Designing one letter and creating an entire font accord

Here is an excellent idea. what if we don't want to use the existing filters? I want a new filter. I can create a new filter using a prompt i.e. prompt-based filter. The user will enter the details of the filter and the ai will generate a filter based on the prompt.

When trying to generate characters from the books I read, their apprerances where completely different from what they're supposed to look like. You should resolve this issue and make sure your AI is capable of reasearching through the internet to correctly know how to depict these characaters, celeberties, animals, or whatever a person who is using your product wants to generate. With the rise of AI capable of generating images from text accurately, more and more people will begin to grow intrested in exploring with it, and for these people, the features they will priotize when searching for what to explore with are going to be speed, intelligence, and accuracy of the AI in responce to the prompt they submit. 

My idea is if you added a way to conver an image type, what i mean is let say i have a portrait selfie photo, and i want to change it to an ID Headshot, or resize it to a passport format, Maybe its a good idea, but i just want to share it with you. FIREFLY IS AMAZING, 

We need PS Beta with AI in the IPad Pro and the full Premier PLEASE naw is the time !

Actually it simple as it is ;)

The AI studies the start (and end) reference images and creates a rough low res 3D representation of the scene (Gaussian splats?). Adobe provides a viewer and tools in which you can guide the camera movement and timing, ease in and ease out, speed, arc, etc.   Then the AI executes the render based on the accurate instructions. Word prompts aren’t enough. Currently the AI is very good at misunderstanding clear instructions and/or making artistically bad choices with the camera movement which wastes credits.

Firefly should produce multiple camera angles of the last frame of video for continuation. The user should be able to select a variation of last frame where the faces and bodies of the main characters are visible to produce the next sequence.

the image editing should have more option to add text. I was trying to make a image with certain description which has some text in speech bubble but AI is not adding the exact text as I mention. I added the text Description: This emoji captures the essence of student life abroad. It depicts a student studying with a book or laptop while wearing headphones, and a speech bubble saying "Struggle is real."Colors: Green for the book, black for the headphones, and orange for the student's outfit.if you see in the image it generated dies not say "Struggle is real" and even colors are not as i mentionedI have attached the image it generated it would be great if we can select this image and just modify some details which we don't like or want. because it always regenerate the whole new image and doesn't allow you to edit just specific details in the image you like

Seria interessante adicionar um recurso em que é possível selecionar uma das opções de imagens gerada e em cima desta opção ter a possibilidades de gerar uma nova imagem usando aquela de referência, assim podendo adicionar comando novos e aperfeiçoar ainda mais a imagem.

It's nice to remove and generate backgrounds, but it seems this lack a few more functionalities:- Outpainting.- Selective inpainting (I've not been able to select a zone in the image to fulfill this).- Document resize.- Negative prompts. Just to name some 🙂

Would be fantastic if Generative Inpaint had an option to resize the image, or basically to just zoom it out a little. Usually, you'd want to some space to add stuff, and without that extra space the process fills cramped.

I feel if you can merge both "TEXT TO IMAGE" and "GENERATIVE FILL" together then we can be more creative and make more customer ai images as our requirement. 

Similar to Adobe PDF Print for any Design Tool, can Firefly be linked to PowerPoint, and other Presentation Tools, which can design concepts and Thematic cover pages with ease. The cover and title of any PowerPoint presentation is important as its the opening page and its the opening slide which sets the context of the presentation and its thought. Making it more visual and logical can add value. Connect Firefly Generative AI to PowerPoint as Add ON plz