|
|
3D video tracking
| Organization: | EPFL (Swiss federal institute of technology Lausanne), 1015, CH | | I.P. Brief: | key point and non-rigid surface tracker for augmented reality applications. The software allows the tracking of 3D objects perceived through a monocular camera without requiring complex and costly stereographic sensors. | | Summary of I.P.: | This software is designed to detect and recognize feature points in images under varying perspective and light condition changes. It is designed so that the detection and recognition is sufficiently fast to be used in real-time applications. Furthermore, it can also compute the 3D pose of a planar object from the recognized keypoints. | | Patent: | na (software) | | Keywords: | 3D video tracking, augmented reality, video games, machine vision, movie industry | | Primary Industry: | Entertainment, manufacturing | | Specific Market: | video games, machine vision, movie industry, entertainment, manufacturing | | Market Size: | Both entertainment and machine vision are huge markets with high growth potential. | | State of the Art: | 3D trackers using sensors | | Figures of Merit: | Only needs a single mongraphic camera, cheap, run on a common PC and webcam, excellent resolution | | Tech. Obstacles: | Industrialization of software needed, to enhance reliability to adress changing conditions. | | Market Obstacles: | find a killer application | | Patent Landscape: | none | | Publications: | J. Pilet, V. Lepetit and P. Fua, Augmenting Deformable Objects in Real-Time, International Symposium on Mixed and Augmented Reality, Vienna, Austria, October 2005
V. Lepetit, P. Lagger and P. Fua, Randomized Trees for Real-Time Keypoint Recognition, Conference on Computer Vision and Pattern Recognition, San Diego, CA, June 2005
J. Pilet, V. Lepetit and P. Fua, Real-Time Non-Rigid Surface Detection, Conference on Computer Vision and Pattern Recognition, San Diego, CA, June 2005, Best Paper Award | | Research Team: | More than 10 researchers of the team of prof. Fua work in the domain of 3D tracking. |
|