Google Unveils Text-To-Video With Imagen Video

Summary:

- Not available to the general public
- Trained by an internal Google dataset made up of 14 million video-text samples and 60 million image-text pairs
- Imagen Video model outputs 16 frames of 3fps footage at 24x48 resolution
- Boosted to generate a final 128-frame video at 1280×768 resolution and 24 frames per second – resulting in 5.3 seconds of high-definition video
- Share multiple examples
- “Video generative models can be used to positively impact society, for example by amplifying and augmenting human creativity”
- Another model released is called Phenaki which I will cover in more detail

Resources:

Google teases its own AI text-to-video model: Imagen Video • The Register

Text-to-image models are dated, text-to-video is in now • The Register

Google’s newest AI generator creates HD video from text prompts | Ars Technica

Google demos two new text-to-video AI systems, focusing on quality and length - The Verge

All these images were generated by Google’s latest text-to-image AI - The Verge

Google's Imagen takes on Meta's Make-A-Video as text-to-video AI models ramp up | VentureBeat

Google answers Meta’s video-generating AI with its own, dubbed Imagen Video | TechCrunch

Google Imagen: AI art but it's videos now | Boing Boing

Google Answers Meta's Video-generating AI with Its Own, Dubbed Imagen Video - Latest Tweet by TechCrunch | 📲 LatestLY

Introducing Make-A-Video: An AI system that generates videos from text

Make-A-Video

Phenaki

Worthwhile Follows:

Lexica