Tiktok owner byteDance has a new AI Video creator you need to look to believe


  • Bytedance shows the new Omnihuman AI video model.
  • Omnihuman transforms a single photo into a lifelike video with full body.
  • The videos show realistic song, speech and movement.

Tiktok’s parent company, bytedance, shows a new AI video creator that can produce live videos of people who speak, sing and move around from a single photograph. The new omnihuman model can bring a picture to life with creepy accurate body movements, facial expressions and movements.

Omnihuman’s breakthrough involved training on more than 18,700 hours of video. AI can now mimic how people move, speak and interact in videos. In particular, this AI can create fully moving figures instead of just animate a face or upper body. This means that a single image can be transformed into a video of someone holding a speech, dance or even playing an instrument.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top