Microsoft's AI will turn still images into videos
The tool will be able to create a video in real time using still images and audio files of a person. Microsoft's new AI tool is also said to be capable of creating everything from facial expressions on still images to well-timed head and lip movements to match a song or speech. The company's researchers attached several examples of the tool to their pilot project. They showed that many believed that the videos created using the tool were real. Microsoft has worked with everything from real people to Mona Lisa stills in this experimental project. However, upon closer inspection, the lip and head movements of the stills appeared to be somewhat robotic and inconsistent in some cases, according to the researchers. They also said that it is possible to easily create a deepfake video of a person using this technology of Microsoft. The researchers said that although there is a possibility of misuse of this technology, it also has several advantages. Microsoft's AI tools can be used to increase educational equity. They added that the tool will also be able to solve communication problems. That is, from sending a video message to someone in the absence of a person in an emergency, it can also provide companionship and therapeutic support to someone as needed.
Microsoft is bringing a new AI tool to turn any still image captured by the camera or hand-drawn into a real video. Recently, Microsoft Research Asia unveiled a new experimental AI tool called Vasa-1. News: Engadget.