MoDB: Database System for Synthesizing Human Motion

Enacting and capturing real motion for all potential scenarios is terribly expensive; hence, there is a great demand to synthetically generate realistic human motion. However, it is a central conceptual challenge in character animation to generate a large sequence of smooth human motion, in a synthetic way. We present a novel, database-centric solution to address this challenge. We propose to generate long, synthetic strips of motion from a database of a small set of real motion captured data, in particular, using various similarity-based "joins" of real motion snippets. This demo will illustrate our system MoDB and showcase the entire process of (a) capturing small real motion data and generating `snippets" (b) encoding snippets it into relational data, and (c) generating synthetic, realistic smooth motion by concatenating snippets based on similarity metrics. In particular, the demo will be interactive consisting of an "avatar"--a graphical creature--that co...

• Type: Conference
• Year: 2005
• Where: ICDE
• Authors: Timothy Edmunds, S. Muthukrishnan, Subarna Sadhukhan, Shinjiro Sueda
• Keywords: Database, ICDE 2005, Real Motion Snippets, Realistic Smooth Motion, Smooth Human-like Motion