Close

Search

Search Menu
Photonics Media Photonics Marketplace Photonics Spectra BioPhotonics EuroPhotonics Vision Spectra Photonics Showcase Photonics ProdSpec Photonics Handbook

Wearable Camera Enables Human Motion Capture

Facebook Twitter LinkedIn Email
Researchers at Tokyo Institute of Technology and Carnegie Mellon University developed a human motion capture system called MonoEye, which consists of a single ultrawide fisheye camera mounted on the user’s chest. The system’s simplicity lends itself to a wide range of applications, in sports, medicine, and entertainment.
MonoEye is based on a single ultra-wide fisheye camera worn on the user's chest, enabling activity capture in everyday life. Courtesy of the Association of Computer Machinery.
MonoEye is based on a single ultrawide fisheye camera worn on the user's chest, enabling activity capture in everyday life. Courtesy of the Association of Computer Machinery.

The system features a 280° field of view and is able to capture the user’s limbs, face, and surrounding environment. To achieve robust multimodal motion capture, the system was designed with three deep neural networks capable of estimating 3D body pose, head pose, and camera pose in real time.

The neural networks have been trained with an extensive synthetic data set consisting of 680,000 renderings of people with a range of body shapes, clothing, actions, background, and lighting conditions, as well as 16,000 frames of photorealistic images.

Due to the domain gap between synthetic and real-world data sets, MonoEye’s developers plan to continue expanding their data set with more photorealistic images to improve accuracy. The researchers said they envision that the chest-mounted camera could be transformed into a small, wearable accessory such as a tie clip, brooch, or sports gear. 

The research was presented at the 33rd Association of Computer Machinery Symposium on User Interface Software and Technology.

Vision-Spectra.com
Oct 2020
GLOSSARY
machine vision
Interpretation of an image of an object or scene through the use of optical noncontact sensing mechanisms for the purpose of obtaining information and/or controlling machines or processes.
lens
A transparent optical component consisting of one or more pieces of optical glass with surfaces so curved (usually spherical) that they serve to converge or diverge the transmitted rays from an object, thus forming a real or virtual image of that object.
fish-eye lens
A type of wide-angle lens that has an angular field above 140° and that exhibits barrel distortion. The most commonly used fish-eye lenses have a field of about 180°, though they are manufactured up to 200°.
Research & Technologymachine visioncomputer visionmotion captureAsia PacificAsia-PacificTokyo Institute of TechnologyCarnegie Mellon Universityneural networkslensfisheyefisheye lensfish-eye lenswide anglewide angle lens

Comments
Submit a Feature Article Submit a Press Release
Terms & Conditions Privacy Policy About Us Contact Us
Facebook Twitter Instagram LinkedIn YouTube RSS
©2020 Photonics Media, 100 West St., Pittsfield, MA, 01201 USA, [email protected]

Photonics Media, Laurin Publishing
x We deliver – right to your inbox. Subscribe FREE to our newsletters.
We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.