基於人類步態之任意角度即時性別辨識

Abstract

本篇論文研究的課題是，如何透過人類步態來進行性別辨識。這是一個十分重要但仍未完全解決的問題。在過程中，我們證明了使用GEI(Gait Energy Image)可以有效地描述從不同角度所觀察到的人類步態。並且，以GEI為特徵，我們透過幾個不同方法，建構了人類步態的角度辨識法，以及性別辨識法。最後透過實驗，顯示了依照我們所提出的方法所建構的系統，可以有效地將即時性別辨識應用於實際狀況中。

In this thesis, we investigate an important but understudied problem, gender classification from human gaits. And we have proved the ability of using GEI (Gait Energy Image) as a representation of human gait for arbitrary view angles. Using GEI as a discriminative feature, we constructed angle classifiers and gender classifiers from different approaches. Experiments have shown that our system achieved a good performance and is able to be applied to real-world application.