指導教授：吳家麟臺灣大學：資訊工程學研究所聞浩凱Wen, Hao-KaiHao-KaiWen2014-11-262018-07-052014-11-262018-07-052014http://ntur.lib.ntu.edu.tw//handle/246246/261511本文中，一個低成本和高效率的系統性自動分析半管運動轉播影片方法被提出來。除了場地顏色覆蓋比率，我們發現玩家區域使用顯著目標偵測機制可面對半管運動影片模糊場景的挑戰。此外，基於現有的MPEG 壓縮的影片原生的運動向量（motion vector）我們提出一種可用於偵測旋轉（spin）事件的新穎且有效的方法。實驗結果表明，該系統能有效在半管影片中識別難以被檢測到的旋轉（spin）和滑行（grind）等事件。In this work, a low-cost and efficient system is proposed to automatically analyze the halfpipe sports broadcasting videos. In addition to the court color ratio information, we find the player region by using salient object detection mechanisms for facing the challenge of motion blurred scenes in HP videos. Besides, a novel and efficient method for detecting the spin event is proposed on the basis of native motion vectors existing in MPEG compressed video. Experimental results show that the proposed system is effective in recognizing the hard-to-be-detected spin and grind events in halfpipe videos.口試委員會審定書i 誌謝ii 摘要iii Abstract iv 1 Introduction 1 1.1 What is a Halfpipe? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 What is a Trick? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.3 How to Learn a Trick . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.4 The Problem of Watching Halfpipe Broadcasting Videos . . . . . . . . . 5 1.5 How to Solve Watching Halfpipe Broadcasting Videos Problems . . . . . 6 1.6 Challenges of Event Detection in Halfpipe Videos . . . . . . . . . . . . . 6 2 Related Work 7 3 Proposed Method 9 3.1 Highest Point Detection and Trick Segmentation . . . . . . . . . . . . . . 9 3.1.1 Noise Elimination . . . . . . . . . . . . . . . . . . . . . . . . . 9 3.1.2 Peak and Valley Detection . . . . . . . . . . . . . . . . . . . . . 10 3.2 Player Detection and Tracking . . . . . . . . . . . . . . . . . . . . . . . 10 3.3 Spin Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3.4 Grind Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 4 Results and Discusion 13 4.1 Cut Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.2 Highest Point Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.3 Spin Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.4 Grind Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 5 Conclusion and Future Work 19 Bibliography 216839965 bytesapplication/pdf論文公開時間：2016/09/03論文使用權限：同意無償授權半管運動影片事件偵測動作辨識thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/261511/1/ntu-103-R01922083-1.pdf