- . OmniHuman turns a single image into a lifelike video. The videos feature realistic singing, talking, and movement.
ByteDance (parent company of TikTok) is demonstrating a new AI video maker that can create vivid videos of people singing, talking, and moving from a single photo. The new OmniHuman can bring an image alive with eerily precise body movements, facial gestures, and expressions. OmniHuman’s breakthrough was achieved by training on over 18,700 hours video. The AI can now mimic the way humans move, talk, and interact with videos. This AI can create characters that are fully animated, rather than only animating the face or upper body. A single image can be transformed into a video showing someone dancing, giving a speech or playing an instrument.
This produces a video that is very realistic, regardless of whether the character comes from a photograph or a stylized painting. Below are some examples.
Omnihuman everywhere
It’s easy for us to imagine that OmniHuman will be a huge hit on TikTok if and when ByteDance makes it available. The company offers an AI video maker named Jimeng, but something like OmniHuman would encourage more people to use TikTok.
ByteDance will not enter the space alone. OpenAI’s Sora is a popular name in the AI video industry. But there are many others, including Pika, Runway and Luma Labs Dream Machine.
ByteDance’s model has many potential uses, including recreating actors from the past for new movies or teaching history through the mouths of historical figures. Even digital avatars used in social media and games could become more realistic by adapting to user input in real-time. OmniHuman, a research project at the moment, is demonstrating its capabilities through ByteDance. This suggests that practical applications will be coming soon. The AI character below may be the face of the next video trend on TikTok.
This AI video maker can make a banana type look realistic and could challenge Sora.
Eric Hal Schwartz has been a freelance writer at TechRadar for more than 15 years. He has covered the intersection of technology and the world. He was the head writer of Voicebot.ai for five years and was at the forefront of reporting on large language models and generative AI. Since then, he has become an expert in the products of generative AI, including OpenAI’s ChatGPT and Anthropic’s Claude. He also knows Google Gemini and all other synthetic media tools. His experience spans print, digital and broadcast media as well as live events. He’s now continuing to tell stories that people want to hear and need to know about the rapidly changing AI space and the impact it has on their lives. Eric is based out of New York City.
More on artificial intelligence
