Questions

the generative fill feature does not work

Adobe Firefly is not aware who is "Imran Khan" the former Prime Minister of Pakistan.

I had chairs for empty stage, ballroom, etc even with ", no chairs" "without chairs" "empty"

Make an option to change the colour of the hair without changing the hairstyle 

Bonjour,J'ai 3 éléments  dans mon prompt (papillon, lavande, herbe) et je souhaite que le papillon sit plus important sur l'image, comment donner un poids à cet élément lorsque je rédige mon prompt ?Merci

Tenho usado bastante o Midjourney e o Dalle3-Bing e consigo criar imagens bastante realistas, mas aqui no Firefly, por mais que eu tente, não consigo. Alguém tem algum exemplo de prompt que permita a criação de cenas realistas? Agradeço qq dica.

I use Firefly for specific purposes, but every time I use to text to image, it gives 4 images that do'nt have the correct parameters set, so they are always unusable. For example, all the images I wan to use must be in portrait mode, photo, but they always start off in 1X1 paint. After it gives that first set of 1X1 images, I always have to change a bunch of settings and start over to get what I want.  I assume Im' getting charged a credit every time this happens. Is there some way to "preset" firefly so even the first set of images I get might be usable instead of a waste?

Idea: Implement stronger object recognition algorithms to accurately identify and fill in missing objects, while maintaining their style and details.Example: Completing a partially obscured person in a group photo would replicate their clothing, body proportion, and facial features realistically.

help me

We’re excited to hear your ideas and look forward to seeing your new feature requests! You may ask, What can you do to increase the chances of being heard?    How will the feature help your workflow?Describe why you want the feature - include the problem you're currently hitting and how this new feature would solve that problem. Before working on a feature, the product team will need to understand how your idea fits in and may even develop a better way of solving your problem.   One request per thread: When creating an Idea or Feature Request thread, please keep requests for features to ONE request per thread. Doing this will allow others to vote on the single, specific feature and not confuse Upvotes amongst a list of feature requests. If an Idea thread has multiple requests in it, you may be asked by a moderator to split the multiple requests up into multiple threads so others can vote.    Enlist support!If you feel strongly about a feature, and you know othe

It's taking too much time and not giving what I want

Just tried using AI and have not used it since OCt. I have not used it for a lot of things only a handful of things but nothing objectionable. Now it it tells me after not using it since OCTOBER that I no longer have access because of community violations. WTH.

Buonasera, ho già spiegato in un mio precedente post nella sezione bugs che capita molto spesso, quando uno FireFly con iPad, che le immagini generate  non vengano salvate correttamente. Nella cartella download trovo solo un link non funzionante. 🤬eh... però tale generazione viene comunque conteggiata con i miei crediti disponibili. Praticamente un furto!!! 🫣Vorrei inoltre segnalare che, se si tenta di salvare l'immagine nella libreria immagini (cliccando su "scarica, poi su "mostra" e quindi sull'immagine per scaricarla, nel ritornare alla schermata precedente tutto sembra congelato e non è possibile fare più nulla finché non si ricarica la pagina.inoltre, trovo irritante che alla prima schermata di FireFly2 occorra inserire forzatamente un prompt senza poter aggiungere un'immagine di riferimento. Praticamente l'applicazione ci costringe a effettuare una generazione inutile che ovviamente viene conteggiata dai crediti. 🤬Inoltre, permettetemi di far notare che i crediti re

Hello, I'd like to generate the following image: a photo I upload is presented as a painting on the wall in a living room. No frame. Firefly generates a picture in a living room, that's OK, but the picture in the frame is not the one I downloaded.Could you tell me if this is possible ?

Does anyone have a working prompt for consistant margins around rendered images? Preferably with definable margins, or at least simply providing a reliable solid margin to avoid the awkward clipping of every image. I've included an example of the desired outcome.

create a image generator like if we uploada image its should show the details and information of image  for example : if a actor is the image and we dont know the name of actor we can find it on image information generator 

Hi there,I'm trying to get a doll with robot head, seems not that hard, but seems impossible for the AI to do this, human "meat" body with tin metal head. Once I write something like "mix" or "with futuristic style" looks like the AI aplies the mixing to everything... I waste a lot of credits trying to get a worthy result.Happy to hear sugestions...The attached image obviously is not what I was looking for._PROMPTGenerate a visually captivating digital artwork featuring a hyperrealistic plastic baby doll body seamlessly integrated with a classic tin robot toy head from the 1960's, situated in a futuristic library setting. The scene should depict the doll engaging in the act of typing on a vintage machine writer. Emphasize the synthesis of organic baby features such as belly, legs, and arms with the robotic elements of the tin head. Convey a harmonious blend of old and new within the library's atmosphere, ensuring the result is a captivating fusion of nostalgia and futuristic technology

Idea:To be able to reference one part of an image, when using generative fill on another. Reason:Generated illustrations can have multiple parts, and sometimes one is bad. E.g. a bad eye in a generated person. When using generative fill, Firefly uses information about the area surrounding the eye, which isn't of much use. It should use the other eye instead. Like the Photoshop Patch tool. Example:One bad eyeball, one good.The “wrong” eye is painted out, for generative fill, and the “proper” eye is selected for reference         A similar eye is generated

Los rostros que generan son deformes y de muy mala calidad 

Hallo Firefly-Team, ich kann meine gespeicherten Favpriten nicht mehr sehen, es werden nur noch graue Felder angezeigt. Wäre toll, wenn ihr mir sagen könntet, woran das liegt.   VG [Removed by Mod]

Hallo Firefly-Team, ich habe versucht, ein Referenzbild, das ich zuvor in Firefly generiert hatte, hochzuladen. Das klappt aber nicht, an Urheberrechten kann es nicht liegen. Was könnte der Grund dafür sein?   VG [Removed by Mod]

There are so many problems, without any reason uploaded picture are hidden.

Devo poter comunicare con l'IA....devo poter dire questo l'hai fatto bene questo l'hai fatto male risolvimi il problema di questo e lui mi deve poter risponde in modo che ci sia un flusso di composizione aggregato...non si può andare continuamente per tentativi... Bisogna creare un interfaccia che renda possibile sviluppare un progetto cioè io do un isiemme di referenze spigo il progetto e poi sviluppo le immagini se non ogni volta si comincia da zero lui non capisce se ne va di testa sua e non si segue il flusso di lavoro... 

Does Firefly Text to image provide illustrator or eps file format to download?

When searching for Korean Hanbok, a lot of dresses show up, but none of them looks like a Korean one. This should be solved asap, there are so many good references online...