利用少量慣性感測器監控人物動作之研究

Abstract

利用從感測器取得的資料來重建動作已經是一個相當普遍的研究技術。在這篇論文當中，我們呈現一個流程圖是藉由綁在使用者四肢跟軀幹上的四到五隻慣性感測器來重建整個人體動作。基於這些收集的資料，我們建立一個包含著十多萬幀線上ｋ元樹的架構並從中取得最適當的動作片斷來決定目前使用者的全身動作。然而由於少量且有雜訊的感測資料通常會造成我們判斷動作的誤差，因此有著發生動作之間不連續的可能性。為了防止這項限制，我們將運動領域概念來避免發生此情況，並能達到更為合體的動作轉換，我們利用即時的動作合成機制將多個動作候補依據其比重關係來混和，使其能夠重現出更為自然且平順的動作，我們的主要目的是利用少量的慣性控制器來達到運用高昂儀器所作出的準確度，並且有著不受環境影響和自我屏蔽的限制。

Motion reconstruction from sensor data is a notable research field. In this thesis, we present a framework to reconstruct full-body human motion by four to five inertial sensors that attached to the user’s four limbs and torso. Based on the gathered data, we construct an online k-dimensional tree (kd-tree) index structure which consists of hundred thousands of frames, and find the most appropriate motion fragment as user’s current full-body motion. However, the sparse and noisy sensing data cause high ambiguity for our motion estimation. It then results in gaps between poses continuous. Consequently, we include the concept of motion fields for more reasonable motion transition. This run-time motion synthesis mechanism merges the candidates of the motion sequences by weighted combination, and generates natural and smooth motions.