Efficient Point-Based Pattern Search in 3D Motion Capture Databases

Abstract

3D motion capture data is a specific type of data arising in the Internet of Things. It is widely used in science and industry for recording the movements of humans, animals, or objects over time. In order to facilitate efficient spatio-temporal access into large 3D motion capture databases collected via internet-of-things technology, we propose an efficient 2-Phase Point-based Trajectory Search Algorithm (2PPTSA) which is built on top of a compact in-memory spatial access method. The 2PPTSA is fundamental to any type of pattern-based investigation and enables fast and scalable point-based pattern search in 3D motion capture databases. Our empirical evaluation shows that the 2PPTSA is able to retrieve the most similar trajectories for a given point-based query pattern in a few milliseconds with a comparatively low number of I/O accesses.

Efficient Indexing of 3D Human Motions

“…Many existing retrieval techniques [2,18,19,24] focus solely on search quality and do not discuss the efficiency at all, which leads to expensive sequential scan over the whole dataset. The efficiencyoriented works either propose very compact features that allow fast sequential scanning [12,13], or utilize various indexing schemes to organize the motion data (e.g., the binary tree [25], kd tree [9], R* tree [4], inverted file index [14], or tries [8]). To optimize the efficiency-effectiveness trade-off, a two-phase retrieval model is often used, where the candidate objects identified within an efficient search phase are submitted to a re-ranking phase that refines the result using more expensive techniques (e.g., traversal of a graph structure [9] or ranking by the Dynamic Time Warping [14,20]).…”

“…A more thorough discussion and comparison of all these methods can be found in the recent survey [21]. However, even the index-based approaches [4,8,9,14,25] are designed to operate on collections of only thousands, or maximally dozens of thousands of motions [22], and their application to large-scale collections is disputable.…”