Facebook Develops AI That Can Help Choreograph Dance Moves to Any Music

Could this be another way for Facebook to take on TikTok?

This week, Facebook has detailed a new AI project it’s developed which is able to outline creative dance moves based on any musical input, in order to help dancers and choreographers come up with new routines.

As explained by Facebook:

“The system analyzes a music track from nearly any genre and, just moments later, cooks up some synchronized moves. The system’s code works by detecting quantifiable similarities in a song at two different points in time, then searching for similar mathematical patterns in a giant matrix of dance move sequences. Since the system’s only computational constraint is ensuring that its movements synchronize with the music, it is able to generate novel dance routines, which human judges have evaluated as highly creative in comparison to other generated dances.”

That sounds like such a Facebook approach to creativity – creating a machine learning system that can do it for you. 

Cool right?

I mean, probably not. Probably, most dancers and choreographers would prefer to actually get a feel for the music and use their own approach. But as Facebook notes, they can get blocked in their creative process, which is where this might help. 

Facebook does also point out that it’s not designed to reduce human input:

“Our dancing agents are meant to augment, not replace, human creations by combining the best of what people and machines each excel at.” 

In a broader sense, Facebook says that the system is significant in that it could help in the development of creative AI: 

“Researchers have made significant progress in teaching machines perception, reasoning, and language in recent years, but teaching AI to be creative and to generate aesthetic attributes is a different open challenge because of how subjective these areas are. […]  The system could someday provide inspiration and creative insights for dancers and choreographers, whether they are amateurs busting some moves in front of the bathroom mirror, or dance industry professionals looking for a new take on a classical ballet production. In the longer run, dancing AI might have potential in video games or fitness apps where people imitate the movements of an avatar.”

So it’s not about coming up with the next hashtag dance trend, as such. But still, it feels like that could be where Facebook is looking, as it seeks out new ways to combat the rise of TikTok, in order to halt the video app’s massive growth.

As noted, creative elements like this have never been Facebook’s strong suit. Facebook’s AR masks and filters, for example, are more technically advanced than those available on Snapchat, but Snapchat is the platform that’s seen continued success with viral trends via its AR tools. Because Snap’s creative team is just better – and while Facebook has sought to open up its AR creation tools to outside creators in the hopes of catching onto the next big trend before it emerges, its own, internal creative approaches have never been overly great, outside of functional developments.

But maybe, AI is the answer to creativity in this sense. Maybe, through sheer resource capacity and tools, Facebook can crack the code of creativity and come up with new, viral dance trends that could bring more people to Facebook.

Facebook has been experimenting with TikTok-like hashtag trends in its main app, and it has ‘Reels‘, its TikTok clone, in Instagram.

The broader view, as Facebook notes, is creativity in AI. But it’s hard to shake the feeling that Facebook has developed this specific tool with a view to tapping into viral trends before TikTok.  

Will that work? Maybe. It’s early days, but it could become a new feature in Facebook to help expand its creative tools.

And if Facebook can expand that to other areas, maybe, Facebook could indeed get a leg up on future creative trends through the use of AI, without having to rely on actual humans.

That, again, would align with Facebook’s approach – but it doesn’t feel like machines are going to be able to facilitate the same level of engagement as human creators. At least, not yet.

You can read more about Facebook’s choreography AI project here.