Google’s artificial intelligence (AI) generator can output 1280x768 high-definition (HD) video at 24 frames per second from a written prompt. The company announced the development of Imagen Video today, which is currently in a research phase. This tool comes five months after Google Imagen, which creates images from text descriptions. Together, these two tools demonstrate the rapid progress of video synthesis models.
To generate the videos, Imagen Video first splits the prompt into sentences and then translates each sentence into a different scene. For example, if the input is “A man wakes up in the morning. He gets out of bed and stretches”, there would be two scenes in the generated video – one where the man is in bed and another where he is stretching. Next, for each scene, Imagen Video extracts relevant information such as objects, actions, and their order from the sentence. Finally, it renders the scene by creating a three-dimensional environment and inserting virtual characters that carry out the identified activities.
There are many potential applications of Imagen Video. For instance, it could be used to create educational videos or to generate demonstrations of how products work. Moreover, as technology develops, it may be used to create realistic computer-generated characters for movies and video games. Importantly, Imagen Video has the potential to democratize video production as it does not require expensive equipment or specialized skill sets. All that is needed is a written description of what you want your video to show.
Google’s newest AI generator creates HD video from text prompts, which has numerous potential applications. This tool demonstrates the progress of video synthesis models and has the potential to democratize video production.
Author: Christian Kromme
First Appeared On: Disruptive Inspiration Daily
The latest disruptive trends with converging technologies that will change your life!