The image below is an artificial intelligence-generated clip from the Vidu website. The tool can create videos based on text or image prompts.
Evelyn Cheng | CNBC
BEIJING — Beijing-based Shengshu Technology said on Wednesday that its artificial intelligence-powered text-to-video tool Vidu will now be able to produce videos by combining images.
Vidu already allows users around the world to create 8-second clips based on written prompts. And OpenAI — Maker of ChatGPT ——Revealed its artificial intelligence model Sora in February A one-minute video can be generated from text, but it has not yet been released publicly.
Vidu’s new artificial intelligence feature can combine three images, such as a shirt, a person and a moped, into a movie of a person wearing a shirt driving a moped through a scene, Moriki said.
Other platforms claim they can use artificial intelligence to convert text or images into video, but the quality of the output varies. Sacred Number’s claimed breakthrough is the ability to take three unique images and integrate them into an AI-generated movie in a visually consistent way.
“We discovered the (visual consistency) problem very early and hope to solve it well,” Fan Bao, chief technology officer of Shengdu, said in Mandarin (translated by CNBC).
Launched in April, Vidu went viral on TikTok by turning two profile photos into realistic videos of people embracing.
According to a CNBC translation, Tang Jiayu, co-founder and CEO of Holy Tree, said in Mandarin that the artificial intelligence video generator has already made money from advertisers, animators and other enterprises. He said usage fees range from 100,000 yuan to 1 million yuan ($13,871 to $138,711) per customer per month.
To address copyright issues, Tang said a company might sign an agreement with an artist to allow artificial intelligence to create ads that mimic the artist’s painting style. He said he has not seen any major legal cases regarding consumer use of the images.
Tang added that Vidu does not allow the public to generate content using images of celebrities or “sensitive” individuals. He said the AI tool also bans nudity and violent images. As for personal photos, Tang said Vidu destroys data in accordance with the General Data Protection Regulation, a global benchmark.
Shengshu was founded last year, with backers including Baidu Ventures, Alibaba’s Ant Group, Chinese startup Zhipu Artificial Intelligence, Qiming Venture Partners and the city of Beijing, according to PitchBook.
Vidu’s artificial intelligence operates on rented cloud servers in China and abroad, Tang said.
