Motion capture on your smartphone with MobilePoser – no suits, no cameras
By Ed Malaker
If you are a game developer or work with animation, you’ll be happy to know there is a great new app available that takes a lot of the work out of motion capture. The app is called MobilePoser. Let’s see what it has to offer.
Benefits of MobilePoser
MobilePoser allows you to perform motion capture with your smartphone without the need for expensive equipment, which is a serious hurdle for independent studios on a tight budget.
Unlike traditional motion capture systems that require suits covered in sensors and high-tech cameras, MobilePoser uses the accelerometers, gyroscopes, and magnetometers already present in devices we commonly use, such as smartphones, smartwatches, and earbuds.
How MobilePoser works
The accelerometers and other instruments in our smartphones can track our position in 3D space. Though they are not precise enough to use for motion capture on their own, the team behind MobilePoser has figured out how to use AI to fill in the gaps. This algorithm refines the data captured by the IMUs, comparing it against high-quality motion data to estimate joint positions, rotation, and even walking speed. Additionally, a physics-based optimizer ensures that movements appear realistic, with constraints like joint flexibility accounted for.
Users can achieve even greater accuracy by wearing multiple devices. For instance, a user might keep a smartphone in their pocket while wearing a smartwatch, and the stem can use many different devices to get the information it needs.
Practical applications and future potential
- MobilePoser opens up new possibilities for those seeking affordable motion capture solutions, especially in gaming, where physical interactions enhance immersion, allowing smaller studios to tackle games in a 3D landscape.
- MobilePooer can help fitness enthusiasts monitor their exercise form to reduce the risk of injury.
- It can help healthcare providers track patient’s mobility without in-person visits.
The developers have open-sourced their models, encouraging collaboration to expand this technology’s capabilities, and it will be available soon on iOS and watchOS platforms.