TikTok owner ByteDance has a new AI video creator you have to see to believe

(Image credit: ByteDance)

ByteDance is showing off the new OmniHuman AI video model.
OmniHuman transforms a single photo into a lifelike, full-body video.
The videos show realistic singing, speaking, and movement.

TikTok parent company, ByteDance, is showing off a new AI video creator that can produce vivid videos of people talking, singing, and moving around from a single photograph. The new OmniHuman model can bring an image to life with eerily accurate body movements, facial expressions, and gestures.

OmniHuman’s breakthrough involved training on more than 18,700 hours of video. The AI can now mimic how humans move, speak, and interact in videos. Notably, this AI can create fully moving characters rather than just animating a face or upper body. That means a single picture can be turned into a video of someone giving a speech, dancing, or even playing an instrument.

The result is a very realistic video, whether the character is a human from a photograph or one from a more stylized painting. You can see examples below.

OmniHuman everywhere

If and when ByteDance does make OmniHuman available, it's easy to imagine it blowing up on TikTok. The company already offers an AI video-maker named Jimeng on the platform, and something like OmniHuman could entice many more people to play with TikTok and its other features.

Of course, ByteDance won't enter the space without competition. OpenAI's Sora has drawn accolades and is a big name in the AI video space, but there are plenty of others, such as Pika, Runway, Pollo, and Luma Labs' Dream Machine.

There's a lot of potential use for ByteDance's model, whether recreating actors of the past for more movies or teaching students history from the simulated mouths of historical figures. Even digital avatars for social media and gaming could become more lifelike, adapting in real-time based on user input.

OmniHuman is still a research project for now, but the fact that ByteDance is already showcasing its capabilities suggests that practical applications aren’t far behind. The AI character below could be the next face of a video trend on TikTok.

See more Computing News

Eric Hal Schwartz is a freelance writer for TechRadar with more than 15 years of experience covering the intersection of the world and technology. For the last five years, he served as head writer for Voicebot.ai and was on the leading edge of reporting on generative AI and large language models. He's since become an expert on the products of generative AI models, such as OpenAI’s ChatGPT, Anthropic’s Claude, Google Gemini, and every other synthetic media tool. His experience runs the gamut of media, including print, digital, broadcast, and live events. Now, he's continuing to tell the stories people want and need to hear about the rapidly evolving AI space and its impact on their lives. Eric is based in New York City.