|
Getty Images & Nvidia. Generative AI: here we come. After months of fighting generative AI model vendors and taking them to court, Getty Images is launching its own generative AI tool that, in contrast to rival text-to-image engines, was trained only on licensed content from Getty's own vast creative collection – but not the editorial images and news photos that contain famous people, brands and other potentially protected content.
This means that Getty's tool won't generate a Nike sneaker because it doesn't know what Nike is, but it can generate an image of a generic sneaker. Getty partnered with Nvidia, which provided the processing power and technical expertise needed to train the model.
Photo Lab. Remini: here we come. While Remini’s AI-selfie effects generator has gotten wide acclaim – and gone viral – past Visual 1st presenter Photo Lab is now launching its LookLab feature to enable users to create multiple AI-generated versions of their photos, showing the subjects in a variety of different poses, angles, facial expressions and surroundings.
What’s different? LookLab only requires uploading a single photo instead of a whole series and lets you experience the results within seconds, without the need to wait for several minutes to receive a “ready for download” notification.
Open AI. DALL-E 3: meet your sister, ChatGPT. 18 months after DALL-E 2 debuted, Open AI is about the ship DALL-E 3 next month, while fighting off other AI image generators that have since seized the limelight (most noticeably Stable Diffusion, Midjourney, and Adobe Firefly).
Using its learnings from ChatGPT, DALL-E 3 is said to be significantly better at understanding the intent of prompts, particularly longer ones, and it should also do better in areas that have tripped up image generators, such as text and hands. Its ChatGPT integration will allow users to hone their request through conversations with the chatbot and receive the result directly within the chat app. I must say, DALL-E 3’s demo images certainly look stunning – but that's what they are, demo images.
Note that artists can now also opt out of having certain or all of their images used to train future OpenAI image generation models.
Finally, DALL-E 3 has new mechanisms to reduce algorithmic bias and improve safety, according to OpenAI. For example, DALL-E 3 will reject requests that ask for an image in the style of living artists or portray public figures.
GotPhoto & NextGen Photo Solutions. Acquisition. GotPhoto, volume photography solution provider and past Visual 1st Best Business Potential Award winner, announces the acquisition of its partner Next Gen Photo Solutions, a graphics company offering services including creating dynamic virtual groups, high-quality graphics, memory mates, and other editing services, such as background removal and touch-ups.
Microsoft. Bringing AI image creativity to your phone’s keyboard. Microsoft’s SwiftKey mobile keyboard app for iOS and Android will not only learn your writing style so you can type faster, the app is also to include AI camera lenses from Snapchat, AI stickers, an AI-powered editor and the ability to create AI images from the app.
The AI stickers can be produced by Bing’s Image Creator, based on your own photos or selfies.
Separately, Microsoft also announced that Bing’s Image Creator will add invisible digital watermarks to all AI-generated images — something it calls Content Credentials. This technology uses cryptographic methods and standards set by the Coalition for Content Provenance and Authenticity (C2PA) to add more transparency around AI images. Adobe, Intel, Sony and others have also joined the C2PA.
YouTube. CapCut: here we come. Google announced YouTube Create, a suite of easy-to-use, free tools to make both Shorts and longer videos. The app will allow you to do things like preview splits, trim your clips and add thousands of stickers, GIFs or songs, and apply various creative effects. With Create, YouTube now has a similar video creation tool as TikTok offers with its sister CapCut app.
YouTube. AI-generated Shorts videos. Google also announced its Dream Screen feature for YouTube Shorts, which will allow you to create an AI-generated video or image background just by typing in what you want to see. Type in something like “a panda drinking coffee,” and then the video image appears on the screen and is ready to enrich the world.
The Shorts platform today is now averaging over 70 billion daily views, up from 50 billion in January. And yes, AI will catapult these numbers further.
Perfect Corp. Adding AI to AR app. Past Visual 1st presenter Perfect Corp. adds an AI selfie generator to its YouCam Perfect iOS AR app. AI Selfi empowers users to transform portrait photos into works of art, offering a selection of 20 distinct artistic styles, ranging from Watercolor, Graffiti, Anime, Manga to Pop Art, and even Van Gogh’s iconic brushwork.
Neurapix. AI-based dynamic batch editing. Based on a training set of 20 images, AI-startup Neurapix can now create “SmartPresets” to batch-edit your photos in the style derived from your own training set photos. Neurapix doesn’t just simply apply the same filter to all images but Neurapix says that they dynamically adapt to the images to make sure they look correct, regardless of subject matter or lighting.
| 
