Most motion-based tracking algorithms assume that objects undergo rigid motion, which is most likely disobeyed in real world. In this paper, we present a novel motion-based tracking framework which makes no such assumptions. Object is represented by

在本文中，我們首先實作了 一個針對壓縮視訊（像是 H.264/AVC）之畫面解析度 改善 的方法。接著，我們分析這個方法的模擬結果。-In this thesis, we first implement a spatial resolution enhancement algorithm for H.264/AVC compressed videos. Based on the analysis of the simulation results, we ide