Data captured by a Tele-Immersion (TI) system can be very large. Compression is usually needed to ensure real-time data transmission. Our compression method takes advantage of prior knowledge of objects, e.g. human figures, in the TI environments and represents their motions using just a few parameters. The main steps of our approach include: motion estimation and residual computation as shown in the figure above. The proposed compression method provides tunable and high compression ratios (from 50:1 to 5000:1) with reasonable reconstruction quality. Moreover, the proposed method can estimate motions from the noisy data captured by our TI system in real time.

See our results: avi (73.3 MB), divx avi (27.2 MB)

Publications

Jyh-Ming Lien and Ruzena Bajcsy. "Skeleton-Based Compression of 3-D Tele-Immersion Data", Proceedings of the ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC), Vienna, Austria, Sep. 2007, to appear.

Jyh-Ming Lien and Ruzena Bajcsy. "Model Driven Compression of 3-D Tele-Immersion Data", Proceedings of the Fifth International Conference on Intelligent Multimedia Computing and Networking (IMAI), Salt Lake City, Jul. 2007, to appear.

See also