© IE Online Media Services Pvt Ltd
Tags:
Journalism of Courage
The software giant Adobe teased as many as eleven new AI-powered ‘prototype’ tools and features on the second day of its MAX conference in Los Angeles on Wednesday. The maker of Photoshop previewed forward-looking technologies, including a tool that brings the power of Firefly generative AI to video for the first time, placing it squarely in the midst of the ongoing AI race.
The biggest news coming from “Sneaks,” an event where Adobe shows experimental features developed by its engineers and research scientists, was the preview of Project Stardust. The tool uses artificial intelligence to recognize key objects within a photo and make editing much simpler. During the on-stage demo, Adobe showed how users can detect unwanted objects in a photo and move or remove them without going through the long process of outlining them. Adobe claimed it would also delete the shadow from the object and replace the missing pixels with realistic AI generation. Some may compare Adobe’s “Project Stardust” with Google’s Magic Editor on Android, but both tools are meant to make photo editing straightforward.
But the most impressive Sneaks demo was Project Fast Fill, a tool that offers the same generative fill technology introduced in Photoshop to video editing, letting users easily remove objects or change backgrounds in videos as they work with a still image with a simple text prompt. Adobe says this even works in very complex scenes with changing lighting conditions.
Another experimental yet powerful AI tool that Adobe showed was Project See Through, and as the name suggests, it removes annoying things like reflections in the glass from the picture, a common problem we all face on an everyday basis. Adobe also previewed Project Draw & Delight. The tool can convert a doodle or sketch into a polished and refined vector drawing with a text prompt. Then there was Project Dub Dub Dub, which can automatically translate a voice recording or video audio track into an unspecified number of supported languages while retaining the original speaker’s voice. Project Primrose also hogged the limelight during the Sneaks. It’s essentially an interactive dress that demonstrates the use case of “flexible textile displays,” letting the wearer display patterns and images on their body, similar to a programmable screen.
In some ways, Adobe offering a sneak peek at far-off and experimental AI tools such as Project Stardust shows the speed at which generative AI is changing creative work. The theme for this year’s MAX conference, a three-day-long get-together that brings people from the creative industry under one roof, is how generative AI can be useful in building upon ideas, rather than replacing professional designers. The company debuted several new AI models, including the next-generation Firefly Image 2 model that produces higher-quality imagery than the initial Firefly model. In addition, Adobe announced the addition of new AI capabilities across Adobe Creative Cloud apps and Adobe Express.
The writer is in Los Angeles on the invite of Adobe.
