Google Lumiere is a video-generation AI model.
It uses a new diffusion model which is known as Space-Time-U-Net.
One differentiator here is that Google Lumiere focuses more on the movement itself.
I am astonished by the growth that we’re seeing in the Generative Artificial Intelligence space. We’ve gotten used to ChatGPT, which generates text based on our written prompts. In recent times, we have seen AI create videos too. ChatGPT has it now and other companies too. Today, we are going to talk about Google Lumiere. It creates videos from the prompts that you type in, and it does it pretty well. Let’s find out more about it.
What is Google Lumiere?
Google has recently launched “Lumiere.” Now, what does it do exactly? Google Lumiere is a video-generation AI model. It uses a new diffusion model which is known as Space-Time-U-Net, or STUNet. This STUNet works on the concept of space and time. So it is basically where things are in the video and how they will eventually move and change. This process allows Lumiere to create the video in a single go instead of going by the traditional route of putting smaller still frames together.
Also read: Google’s Bard chatbot now available in Gmail, Docs, Drive, Maps & more: Here’s how it works
How does it work?
Let’s now talk about how the process works after you have entered your prompt. Once you give Google Lumiere the prompt, it then uses STUNet framework and figures out where the objects that you have asked for will be placed in the frame and how they will move to create more frames that eventually flow together. This eventually allows the video to appear in a more seamless motion. One differentiator here is that Google Lumiere focuses more on the movement itself and where the content that it has generated should be at a given point. Other models usually stitch the video from generated keyframes where the movement has already taken place.
Also read: Google Chrome to get new generative AI features: Tab Organizer, custom themes & more
Google Lumiere also generates 80 frames instead of 25 frames from Stable Video Diffusion. This helps the videos to appear nearly realistic, as is evident from the reel that Google published. However, since it is AI, it does show us glimpses where the video feels artificial. But it is still impressive.
Would you use Lumiere for your video creation needs? Let me know in the comments below.
Follow Us
Mustafa Khan
Mustafa is new on the block and is a tech geek who is currently working with Digit as a News Writer. He tests the new gadgets that come on board and writes for the news desk. He has found his way with words and you can count on him when in need of tech advice. No judgement. He is based out of Delhi, he’s your person for good photos, good food recommendations, and to know about anything GenZ. View Full Profile
