Strange IndiaStrange India

Adobe has announced its own family of creative generative AI models, Adobe Firefly. The San Jose, California-based software company has announced the beta launch of a family of creative generative AI models. At launch, the company has released the first two tools that are part of the family of creative generative AI models. One of the tools is a text-to-image AI generator that works much like DALL-E and Midjourney, while the second tool is a stylised text generator that works much like the AI-powered WordArt launched by Microsoft.The Photoshop fame software company Adobe announced its own family of AI generative tools called Adobe Firefly through a dedicated website. The company is currently calling its latest AI tool a beta version and is inviting users to be part of the beta program by signing up on the official website. However, Adobe plans to make the tool available as part of its popular suite of creative apps like Photoshop, Illustrator, and Premiere, in the near future.As part of the company’s AI-powered illustration, artwork, and graphic design tools, users will be able to create custom vectors, brushes, and textures using text or sample sketches. This will allow users to create new tools and edit images and videos using them. Adobe Firefly also includes AI-powered 3D modelling features that will allow users to turn 3D compositions into photorealistic images using just a few steps.Adobe also claims to be open about the datasets being used to train its AI models. The company claims that all of what is being fed into its generative AI models is either copyrighted, licensed for training, or from the Adobe Stock Library, claims Alexandru Costin, VP of generative AI and Sensei at Adobe, according to a report by TheVerge.“We’re not afraid of change, and we’re embracing this change. We’re bringing these capabilities right into [our] products so [customers] don’t need to know if it’s generative or not,” added Alexandru Costin, according to the report.Adobe also plans to introduce a Photoshop feature that would allow users or artists to train the system on their own work in order for the system to help generate content using the artist’s signature style.The company is also reportedly working on a ‘Do Not Train’ system which will allow artists to embed a request into an image’s metadata that would prevent it from being used to train AI systems in the future, as long the creators oblige to the request.Adobe Firefly’s first two tools are available for public beta testing starting today and users wishing to test them out do not need to be a Creative Cloud subscriber to request access. However, Adobe will be limiting the number of users who will receive access to the AI tools through the beta program.From smartphones with rollable displays or liquid cooling, to compact AR glasses and handsets that can be repaired easily by their owners, we discuss the best devices we’ve seen at MWC 2023 on Orbital, the Gadgets 360 podcast. Orbital is available on Spotify, Gaana, JioSaavn, Google Podcasts, Apple Podcasts, Amazon Music and wherever you get your podcasts.Affiliate links may be automatically generated – see our ethics statement for details.

Source link


Leave a Reply

Your email address will not be published. Required fields are marked *