Feature Requests

Me ha gastado cuatro imagenes, cobrandome créditos y no sacando las imagenes originales, sino cópias. Quiero poder hacer el vídeo que le pedí sin pagar más créditos!!!!!

Das Foto das ich von Lightroom zum bearbeiten nach Firefly sende wird dort nicht angezeigt

Sería genial que pudiéramos retroalimentar en el chat lo que queremos que la AI haga para nosotros, de modo que podamos enseñar a la AI la manera en que nosotros seguimos ciertas instrucciones, para conseguir el objetivo de diseño. Por ejemplo, poder explicarle conceptos sobre un tipo de dibujo y que esta logre comprender de forma básica los fundamentos especiales para completar nuestro diseño. Otro ejemplo es, si quisiera corregir un error que cometió la AI, quisiera poder señalárselo inmediatamente de forma que la retroalimentación sea tal cual. Poder enseñar a la AI a tener estrictamente los parametros que le quiero imponer, ayudaría bastante a optimizar las tareas de los jugadores y demás clientes. 

I'm sure you already have it on your radar, but it would be very helpful for users to have a loop feature available, especially since we are currently limited to 5 seconds of generation right now.  Like I said, I'm sure you have it in the queue, but part of the reason I'm generating so many generations is that I am exporting the last frame of a generation, using that as the starting image for my next video in the sequence, trying to string together multiple 5 sec generations to complete one video. I know this will resolve itself as features are released and unlocked.   Thanks again, Paul

Maybe I am just new to using generative text-to-image tools. Still, It turns out that getting Firefly (or DALL-E, and several others I tried) to pay attention to and follow instructions that specify or adjust the direction in which a 2D artwork subject is oriented/is facing/is gazing. This may be easier to accomplish with photography subjects, with better face detection in photos. Not so for cartoon bunnies staring dramatically in the wrong direction. It was particularly difficult (as in tens of hours of trial and error) to get Firefly to reposition a small, AI-generated drawing of a bunny (attached) with a yaw of about 45 degrees (45 degrees counterclockwise away from the viewer) to face in any other direction.  Even more frustrating,  bunny's position was such that his left eye was completely obscured due to the shape and position of his head, and it took hours of trial and error--and frustrated attempts at using puppet warp, liquefy, free transform, etc.--for Fir

Please make provision for uploading multiple source images for video compositing, like Kling "Elements".

Will there be future updates to enhance old photos into videos, similar to those on the MyHeritage website? The current version of Firefly distorts the images too much

I think it would be awesome if we could make adjustments in adobe inpaint and then using that same image we made those adjustments as a base in the text to image. 

create audiobook with my novels.I would prefer the voiceover application to have a Turkish language option.

Its a good idea to in generative fill have a option to upload an image to take element from one image and add to another or composite a image or Ai can look for create similar thing on current image from the uploaded image

Prezados Desenvolvedores,Gostaria de compartilhar uma ideia que considero de grande relevância para o aprimoramento da vossa plataforma. Recentemente, tive a oportunidade de observar as capacidades oferecidas pela tecnologia de processamento de imagem (Adobe Firefly), e acredito que podem elevar significativamente a experiência dos usuários ao implementar opções avançadas de edição de imagem impulsionadas pela IA.Haveria a possibilidade de fazer upload de imagens e, com base nessas imagens, utilizar a inteligência artificial para realizar a criação de cartoons, caricaturas, correção de fotos embaçadas, remover ruídos, figuras semelhantes e até mesmo realizar uma clonagem de fotos, tudo isso com uma qualidade excepcionalmente alta e uma IA mais eficiente em relação à fidelização da imagem.Tendo em mente a excelência associada ao nome da ADOBE, imagino um recurso que se destaque pelo seu nível superior de qualidade. Um exemplo que me inspira é o trabalho feito pela Adobe, reconhecid

I lose too many credits doing trial and error. The monthly cost should be based on downloads

There should be a feature in which we provide 3-10 images and ai will then generate more images like those. For example I am making phone wrap designs and i am out of creativity now, so, this feature can make a number of images and help many like us

Hello Adobe team,   I'm [name removed], thank you for your devotion to AIGC first.   While using the recolor function, I found it hard to get the expected results by word prompting. And I wonder if you may open access to let users upload color references with pic.   This will certainly help users get what they truly need!   Thanks again! And have a nice day anyone seeing this message.    

First of all thank you very much for reading me... And thank you very much for developing this great tool for creative progress... - I apologize because my ideas are just part of my imagination and I don't know anything about programming, so I don't know how complicated it is to carry them out...- But here I leave them wishing to contribute in all this great work..Things you should integrate:.- Possibility to see the history of searches and images.- To be able to show many images (instead of only 4).- Ability to go back to previous samples- Possibility to enable restrictions, especially for cases where people would search for sexual, violent or negative content (This also applies in case of a little boy interacting with the artificial intelligence).- Perhaps the possibility of implementing a basic drawing that allows the intelligence to identify where certain elements go.

It would be great to be able to rearrange and remap filters using a node GUI. 

Hey there, I'd love the opportunity to add own pictures to the generative fill. Let's say I've created a picture with a glass in it with Adobe Firefly / or I just have one. Now I wanna add myself, from another picture I've taken, into the glass - keeping the look of being trapped in the glass though - currently that's a lot of editing to get there, as just showing the underlying layer won't give me that effect. This is a very specific, weird example - but any composit with a texture to it, would benefit from that possibility.  Thanks!

$$.  I’d like to make month to month bill tracker for people who’s memory or need to have it written down to help them understand and keep track of there bills and where there money is going 

Please add Polish as a supported language for the Translate Video feature in Adobe Firefly.Polish is spoken by over 35 million native speakers and is one of the largest languages in Europe, yet it is currently unavailable in the translation feature. This significantly limits the usefulness of Firefly for creators, educators, developers, businesses, and content creators who produce or consume Polish-language content.A few common use cases:    ●    Translating English technical tutorials into Polish while preserving the original audio experience.    ●    Creating localized educational and training videos for Polish audiences.    ●    Making international content more accessible without requiring subtitles.    ●    Helping creators reach a much larger audience through AI-generated voice translation.The current translation quality for other supported languages is very impressive, and adding Polish would make Firefly even more valuable for users in Poland and for the global Polish-speaking

Integrate Chat GPT in Adobe FireFly for Ideation and story telling.

When using the "text to image" functionality, images often contain text and the text often contains spelling errors. Having the option to omit the text could be really beneficial, e.g. a checkbox or specifying a promt like "Create an image representing X where no text is shown on the image".

For example, current might lack knowledge on certain subjects (e.g creating images of Chinese food), but there is no way to teach the AI how to create those images without much trial and error. What if we have a feature that allow us to use one (or more) sample to educate AI (e.g use image A, B, C as sample for image related to item X)

when we select the background allow the front-end picture to move within its frame size to customize corner

For some clients, I may do a simple black an white rendering. Some details and placement of objects must be specific so that an AI that will take a black and white ink drawing and make a realistic image for a client to visualize. As it stand now, I am at the whim of  how the AI decides to lay things out. Maybe we can take a rendered image, do a sketch on top of it and have the AI adapt to the redrawn paramaters? As a mural painter, the creating time will be woefully cut down and the client has a better experience and better representation of the final mural.

By typing shirt or something else the AI automatically select that portion automatically so that the manual selection process become faster and it's become more user friendly