Google introduces LUMIERE, a video generation model, here’s how it works and its feature launch timing
Google Lumiere: tech joint Google has introduced the LUMIERE AI model with the help of which you can create videos by writing text. The company’s new LUMIERE is a text to video and image to video model. This means that you can not only create a video directly from text but can also create a motion video from an image. A video of how this tool works has been shared on X. For your convenience, we are adding it here in the article. With this you can understand better.
How does the LUMIERE model work?
Unlike traditional video models, Google’s new LUMIERE adopts a space-time U-Net architecture that produces the entire temporal span of a video (the entire video by frames) in a single pass. This innovative technique eliminates the need to synthesize distant keyframes after temporal super-resolution, making global temporal stability easily achieved.
According to the researchers, the text-to-video generation framework has been introduced using pre-trained text-to-image diffusion. Because existing methods struggle with globally consistent speed, the team generated full frame video clips by deploying a space-time U-Net architecture that incorporates spatial and temporal modules. Due to the hard work of the researchers, good results have been achieved in Image to Video, Video Inpainting and Stylized Generation.
Special features of this model
- You can create video by writing text
- Can make video from any photo
- Stylized generation, i.e. you can create a photo in its style by taking reference of a photo.
- Can animate a photo
- This model will help you in video editing, like you can change the color of any object, dress etc. with the help of AI.
When will it be launched?
At present there is no news regarding the launch of this model of Google. Maybe the company will introduce it with its bard.