Adobe has released its own neural network for generating short videos: Generate Video is only available in beta for now
Adobe - the developer of Photoshop and other popular graphics software - has released its own neural network for generating short videos.
Here's What We Know
The novelty has a simple name - Generate Video and is similar to Sora from OpenAI and other AI-tools for working with video.
With Generate Video you can create a video up to 5 seconds long (Sora from OpenAI generates videos up to 20 seconds long) with a resolution of 1080p at 24 FPS. It takes about 90 seconds to create a single video.
At the moment, only the beta version of Generate Video is available, which supports three use cases:
- Generate Video from a text description;
- generating video from an image;
- video translation.
Generate Video is part of Adobe's Firefly service and requires a paid subscription to access the neural network.
Users of the basic plan ($10) can generate 20 videos per month, while the Pro version of the subscription ($30 dollars) allows you to create up to 70 videos. The limits will change at a later date.
Adobe is reportedly working on other variants of Neural Network, including accelerated video generation but reduced resolution, as well as 4K video generation.
Source: Adobe blog
