In today’s AI-and-music news, this one can turn you into a dancing fool using just a single still picture
AI continues to evolve in interesting ways. Now we have the Disentangled Control for Referring Human Dance Generation in Real World (DisCo). It can create a dance video of you starting with nothing more than a static picture. This comes as a result of a collaboration between researchers and Microsoft and Tan Wang’s team of researchers from Singapore’s Nanyang Technological University.
It works like this. DisCo splits your image into three parts: your pose, foreground, and background. It then draws from dance moves from TikTok fed into it. Those moves are then applied to your picture.
I thought about trying this with a picture of me, but no one needs to see that. Instead, I offer you this demo.
