Feature Requests

Es wäre schön, wenn man schon so tolle KI-Bilder machen kann, dass die Qualität Druckqualität hätte...Außerdem treten immer noch Fehle auf, wie fehlende Stuhlbeine, oder die Zusammenführung meherer Objekt, die gar nicht so gewollt sind. die 3. Generation istr schon viel besser geworden, aber noch weit davon entfernt, umwerfend zu sein.Ichn hoffe, dass auch eine Video-KI, und sogar eine KI entwickelt wird, die zu meiner Klaviermusik die entsprechende Animation machen könnte, die ich per Text eingebe..z.B. Flug durch einen Wald und über Wiesen, über Flüsse und Bäche, durch ein Höhlensystem und über Berge und Täler... im Rhytmus der Musik

Detailed cross section of the human body with organs *** Please activate and train Adobe Firefly AI for such medical views   generated - not in real picture - just in drawn form! including gender-specific characteristics to generate for medical/explanatory purposes with display, for example, only of certain muscle groups of the female-reproductive organs   Thank you Adobe-Team!

I would love to be able to generate a banner-sized image as I often create graphics to place at the top of documents, websites, web headers, Slack canvas headers, and normally the widescreen content is too tall to really create anything useable. I know I could create content from Photoshop but I love the ease and speed of generating right in Adobe Firefly. Starting with a thinner but wide aspect ratio would help me create useable content very quickly for these other uses. 

Un Lazo Inteligente o una Herramienta de Selección de Objetos, similar a la de Photoshop, sería esencial para el Relleno Generativo en Adobe Firefly en su página oficial. Utilizar un pincel para las selecciones puede resultar tedioso, especialmente si se busca un resultado preciso.  

There is no option to change the hairstyle. I tried many times but it didn't work.

I was thinking it would be useful to be able to convert the AI jpg into a layered psd file to further edit and alter AI images. This would be of value to also take these files and animate them in AE. The AI generated images are a great start but very limited with a flattened image.

When I am looking for white people the results are always someone who isn't and is just the peruvian stereotype. Some results has this bias on my country. This was my prompt: streetwear men clothing, brutalism, gray and minimalist, white people

... it is really a big overlay smack in the middle of the bottom screen, always covering a part of the picture while I zoom in/outCan't it at least be moved to the side or even better, be moved and resized by the user?Also: pan-ing the picture loses the currently generated choices... which is a pain, since I often want to move the pic to better see what it is inserting... and thus need to fiddle with the fickle zooming 😉 ... but all in all I'm pleasantly surprised by the functionality... just started with StableDiffusion Generations a month ago, and this is certainly more restrictive in other users toolings, but it works amazingly well with what I have tried so far 🙂 /kudosThe restrictions on keywords are a bit harsh though, I think... depicting a spacefaring image for example on basis of the topic usually includes lasers and rifles and such - which don't need to be used to depict violence. 😉

Seria bueno generar texto o etiquetas en el generador de imagenes de firefly de una vez si fuera posible como una opcion 

For editing multiple photo the background we getting so random for each photo background is random and different soa person who is already fix a background but in next picture the background that background can be save 

About　FireflyI would like to be able to easily create different poses with the same outfit, different poses with different outfits, or different poses with different outfits from a single photo or illustration.　I want to be able to set facial expressions in detail.

How bout when returning to gallery the gallery page returns to where it was exited. It's getting really old having to start all over each time one returns to the gallery view page. Makes it almost impossible to explore images.

Hi I was wondering if you'll ever create an Ai text to video (Movies)and text to Game

It's tedious to enter the same topics over and over again when I start firefly - on the start screen I can only enter the prompt but no other information... that's a time killer because you have to wait for the generation. Why should the input be structured using different text fields? The AI should be able to recognise easier the meaning and display in a targeted manner. It is also an ux issue for those who have to enter data. If I often prompt different subjects, I don't always want to write background/foreground or prompt forever until a thing appears where the exclusions are not included. Suggestion:Individual text fields for the prompt: "Background:" "Foreground" "Exclude"Enter the format immediately, as well as the effect styles (I also don't like it when you get used to effects and then they are suddenly no longer available).It should be possible to mark/mark areas to indicate things that you like directly in the generation or that you can only further prompt certain a

How about instead of waiting a month for a refill on generative credits, people can get more by rating firefly images? Maybe for every 5 images someone rates, he or she gets another credit. This would be very helpful to us users so we can create more images, and also helpful to you because more images are being reviewed and rated. Thanks for reading!

We have a requirement of keep track of the promts that we use to generate images (legal requirement). This would be so much easier if the prompt was captured in the images' EXIF data. We could then extract the data in PS or Bridge. We could even filter & search in Bridge on the data in the prompt.

I feel like you guys should work on a command that works with the color wheel or color palette of an image. For example, there's a patch of the image I like so I put it into generative fill so I can make that patch match the color I want. I was thinking the user can write something like " I want a darker shade" and it gets filled in. or they can choose the color they want if the command doesn't fix what they want.

Would be nice to have feature to choose solid black background, or solid any color specified background.I find it requires lots of specific prompts to get objects in center on solid black with no floor , no ground plane.

c'est dommage, qu'il n'y ai pas une fonction de "recherche" dans la gallerie

Give option for change dress from gallery 

Create an option to develop things specifically for computer or phone backgrounds. Add a section to implement text for "motivational backgrounds". Or allow users to build on top of other background ideas. Add sections related to the types of backgrounds someone might want such as nature, minimalistic, artistic, etc. 

A dark background would help.

The process of generating images curating to favorite the best options becomes laborious and boring, even more risky that you lose everything if someone deletes cookies from your browser by accident. We urgently need a simple button to download all images from favorites at once or in batch. I would not like to use the save option in the CC library this is a little useless in the usability of the user who always sees 4 result options.PS: I can like images from the community but it doesn't make sense if these likes don't go somewhere and I can consult later, as well as the basics on all the sites that you like images (freepik envato...)

The AI seem not to understand "no background" or "plain white background". Try to teach it this.

Je me sert régulièrement des mes créations en arts plastique tels que peintures, dessins et autres créations prises en photos mais aussi de mes créations numériques notament réalisées avec photoshop et d'autres logiciel. Très régulièrement le logiciel Adobe Firefly ne m'autorise pas à me servir de mes crétions sous prétexte que cela n'est pas autorisé par les standards de la communauté. De plus certains mots dans les prompts me sont bloqués régulièrement pour les mêmes motifs. Si ce logiciel doit être un outil libre de créations ne peut-on pas avoir un bouton ou une case à cocher qui permet d'indiquer, comme une déclaration sur l'honneur numérique, que l'on assume juridiquement la paternité des créations utilisées ainsi que les recherches graphiques associées à nos idées ? Je fais souvent des créations mystiques sur fonds de poésies noires et de spiritualités qui inclus des questionnement sur la vie et sur la mort que je cherche à mettre en images et cela m'est interdit pour le moment