A Chinese AI startup Introduced an Image-to-video Converter

Shengshu Technology, located in Beijing, recently announced that Vidu, an artificial intelligence-powered text-to-video tool, will soon be able to make videos from numerous images.

Vidu currently enables users worldwide to create 8-second videos based on written prompts. 

Although OpenAI — the company behind ChatGPT — claimed in February that its AI model Sora could produce one-minute videos from text, it has yet to make this public.

Shengshu said that Vidu's new AI capability can integrate three images — like a shirt, a person, and a moped — to create a video of the person wearing the shirt and driving the moped around the scenario.

Chinese AI startup launches image-to-video converter

Other platforms claim to be able to convert text or images into videos using AI. However, the quality of the results varies. 

Shengshu claims to have made a breakthrough by combining three distinct photos and integrating them with visual consistency into an AI-generated film.

 Jiayu Tang, Shengshu's co-founder and CEO, stated that Vidu does not enable the public to create content utilizing images of celebrities or "sensitive" individuals. 

He added that the AI program also prohibits naked and violent imagery. 

Furthermore, Tang said that Vidu destroys personal images in compliance with general data protection regulations, which serve as a global benchmark.

Read more: Heroic Shorts Launches an AI-powered Text-to-video Platform