A Mixed Reality Real-Time Hand Tracking on Mobile Devices

• Main Authors: CHAN, YUAN-CHING, 詹媛晶
• Other Authors: WANG, LUNG-JEN
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/87488x

碩士 === 國立屏東大學 === 資訊工程學系碩士班 === 106 === 　　With the rapid development of digital technology, smart-living applications are getting ra-ther important. Facial expressions, body movements, and voice are commonly used for people to communicate with each other and are also expression ways in machine-human interfaces. Among body-movement expressions, hand poses are one of the simplest ways. 　　In this thesis, we propose to develop an application which can immediately track human hand in the sequential images shot by cameras on mobile devices. Tracked hands can be placed in different backgrounds to get visual effect of AR (Augmented Reality) and VR (Virtual Reali-ty). If one needs help, the other can help with doing some hand poses in front of the smart phone camera. Then, both of them can see the combined image composed by scene from the one needs help and images of hands from the help-giving one. Skin detection, illumination correction, and noise removal are used in our proposed application. Related experiments are also implemented to help improve the accuracy. A preliminary success of our proposed application is shown in this thesis.