Basic Search / Detailed Display

Author: 鄭雅勻
Ya-Yun Jheng
Thesis Title: 以手勢玩行動裝置虛擬實境遊戲
Mobile Device Virtual Reality Game Played with Gestures
Advisor: 洪西進
Shi-Jinn Horng
Committee: 李正吉
Cheng-Chi Lee
楊昌彪
Chang-Biau Yang
楊竹星
Chu-Sing Yang
林韋宏
Wei-Hung Lin
Degree: 碩士
Master
Department: 電資學院 - 資訊工程系
Department of Computer Science and Information Engineering
Thesis Publication Year: 2021
Graduation Academic Year: 109
Language: 中文
Pages: 61
Keywords (in Chinese): 深度學習虛擬實境行動裝置手勢辨識
Keywords (in other languages): Deep Learning, Virtual Reality, Mobile Device, Gesture Recognition
Reference times: Clicks: 329Downloads: 0
Share:
School Collection Retrieve National Library Collection Retrieve Error Report
  •   要將虛擬實境(Virtual Reality, VR)推廣至更多人使用,勢必得舒緩使用虛擬實境時會遭遇之限制。使虛擬實境使用族群增加緩慢的重要因素在於運行虛擬實境時,常需搭配其他硬體設備,追求即時的互動與真實的體驗。例如:利用紅外線感測器定位玩家手部位置並偵測其動作。為了搭配這些硬體設備並顯示擬真的畫面,運行虛擬實境的設備需求也會隨之提高,這不僅是增加了使用者的負擔,更導致想體驗虛擬實境領域的使用者卻步。
      為舒緩上述限制,本研究之目標是讓使用者能使用最低的硬體需求即可體驗真實的虛擬實境。本研究提出HandVR用來解決虛擬實境常需搭配額外硬體設備之限制的方案。首先,為取代虛擬實境玩家手部定位與動作辨識之硬體設備,本研究使用MediaPipe Hands深度學習(Deep Learning)模型作為取得玩家手部資訊之來源。此模型以RGB圖像作為輸入,經過運算後,輸出手部共21個關鍵點之三軸座標,再以此結果計算各個手部關節點角度,並模擬雙手在虛擬實境中的姿態與手勢;再來,為同時減緩運行平台之負擔,本研究除了使用輕量化版本的深度學習模型外,也使用先檢測手掌、再進一步分析手部關鍵點的方式,以減輕運算的需求與負擔。HandVR以上述兩大技術為重點,實現僅需使用手機即可運行、使用雙手操作之虛擬實境遊戲。
      本研究結合MediaPipe Hands深度學習模型與Google Cardboard頭戴顯示器,並使用Unity做為開發用的遊戲引擎,設計了一款能在Android手機運行的虛擬實境、且不須架設伺服器之遊戲──HandVR,遊戲中包含了記憶力遊戲與休閒大廳兩種關卡類型,能夠讓使用者體驗手勢辨識結合虛擬實境之特色;HandVR驗證了可行性與可玩性之價值,而在達到大幅降低使用虛擬實境門檻的同時,達成了本研究欲推廣虛擬實境之目的。


    As for promotion of virtual reality, it needs to alleviate the restrictions encountered when using VR. Among those restrictions, the important factor that slows down the increase in the use of virtual reality is when running virtual reality, additional hardware is needed to pursue real time interaction and real experience. For example, infrared sensors are used to locate the player’s hand position and detect its movements; however, in order to match these hardware devices and display immersive scenes, the demand for VR devices will also increase. This not only increases the burden on users, but also makes users who want to experience the VR deterred.
    In order to alleviate the above limitation, the goal of this study is to allow users to have a reality experience with the lowest hardware requirements. HandVR is a solution proposed in this thesis to remove the additional hardware devices. First of all, to replace the hardware equipment commonly used in VR for capturing player's hand position and motion recognition, we use MediaPipe Hands deep learning model which takes RGB images as input, and outputs the three-axis coordinates of a total of 21 key points on the hand for obtaining player hand information, and then calculates the angle of each hand joint point, and simulates the postures and gestures of both hands in VR. Second, to reduce the computational burden at the same time, in addition to using a lightweight version of the deep learning model, we also use the method of first detecting the palm and then further analyzing the key points of the hand to reduce the need and burden of computing. HandVR focuses on the above two technologies, and achieves VR games that can be run only by using a mobile phone and operated by hands.
    This study combines the MediaPipe Hands deep learning model with the Google Cardboard head-mounted display, and uses Unity as the game engine for developing and designing a VR game that can run on Android phones without setting up a server: HandVR. The game contains two types of levels: memory games and leisure halls, allowing users to experience gesture recognition combined with the characteristics of VR. HandVR has verified the value of this study in potential to be promoted, feasibility, and playability. While achieving a significant reduction in the demands for using VR, it also achieves the purpose of this study to promote virtual reality.

    中文摘要 II ABSTRACT III 誌謝 IV 第一章、緒論 1 1.1 研究背景與動機 1 1.2 相關研究 2 1.3 研究簡介 3 第二章、各領域介紹 4 2.1 深度學習介紹 4 2.1.1 架構 4 2.1.2 常見應用 7 2.1.3 深度學習小結 8 2.2 手勢辨識研究 8 2.2.1 介紹 8 2.2.2 工具 9 2.2.3 應用 14 2.3 遊戲引擎 14 2.3.1 介紹 14 2.3.2 Unity Engine 14 2.3.3 Unreal Engine 15 2.3.4 比較 15 2.3.5 遊戲引擎小結 16 2.4 虛擬實境 16 2.4.1 介紹 16 2.4.2 比較 17 2.4.3 虛擬實境小結 18 第三章、研究設計與方法 20 3.1 使用工具 20 3.2 研究設計 21 3.2.1 異步執行 21 3.2.2 手部關鍵點辨識 22 3.2.3 手勢定義 23 3.2.4 遊戲架構 27 第四章、效能評估與成果比較 39 4.1 效能評估 39 4.2 成果比較 41 第五章、結論與未來展望 44 中文參考文獻 45 英文參考文獻 48

    中文參考文獻
    [1] Vive, 找到適合您的高階VR裝置| VIVE 台灣, Vive, Jul. 15, 2021. Accessed on: Jul. 15, 2021. [Online]. Available: https://www.vive.com/tw/product/
    [2] Axis, Manus VR | VR虛擬手套 | 愛迪斯科技, Axis, Jul. 15, 2021. Accessed on: Jul. 15, 2021. [Online]. Available: https://www.axis3d.com/manus-vr
    [4] 王根顥, "基於深度學習的「拿了就走」無人商店," 碩士, 資訊工程系, 國立臺灣科技大學, 台北市, 2020. Accessed on: Jul. 13, 2021. [Online]. Available: https://hdl.handle.net/11296/3mr5ds
    [5] 廖崇儒, "可供輪椅使用者使用之基於電腦視覺的手勢辨識系統," 碩士, 電機工程系, 國立臺灣科技大學, 台北市, 2010. Accessed on: Jul. 15, 2021. [Online]. Available: https://hdl.handle.net/11296/c3vvp9
    [6] 張晉豪, "基於皮膚顏色及手指角度之手勢辨識," 碩士, 電機工程學系, 國立暨南國際大學, 南投縣, 2020. Accessed on: Jul. 15, 2021. [Online]. Available: https://hdl.handle.net/11296/byn3xt
    [7] 曾士宏, "即時手勢辨識系統之應用," 碩士, 電機工程學系碩士在職專班, 淡江大學, 新北市, 2009. Accessed on: Jul. 15, 2021. [Online]. Available: https://hdl.handle.net/11296/86qv5z
    [8] 賴韻芊, "基於電腦視覺之手勢辨識系統," 碩士, 資訊網路與多媒體研究所, 國立臺灣大學, 台北市, 2009. Accessed on: Jul. 15, 2021. [Online]. Available: https://hdl.handle.net/11296/yxhrj5
    [11] 林鈺皓, "使用深度感測器之靜態手勢比對用於門禁系統," 碩士, 資訊工程系研究所, 國立臺北科技大學, 台北市, 2015. Accessed on: Jul. 13, 2021. [Online]. Available: https://hdl.handle.net/11296/nk42d4
    [19] H.-E. Sun, 卷積神經網路 (CNN) 的發展, Dec. 31, 2019. Accessed on: Jul. 13, 2021. [Online]. Available: https://medium.com/ai-academy-taiwan/%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E8%B7%AF-cnn-%E7%9A%84%E7%99%BC%E5%B1%95-4c5d29e60c55
    [33] 李嘉和, 虛擬教育應用服務, 工業技術研究院, Jul. 25, 2017. Accessed on: Jul. 15, 2021. [Online]. Available: https://ictjournal.itri.org.tw/Content/Messagess/contents.aspx?&MmmID=654304432122064271&CatID=654313611331661503&MSID=745167162107206700
    [34] 蘇宇楨, "運用深度學習辨識手語手勢之創新研究," 碩士, 資訊管理學研究所(含碩專班), 世新大學, 臺北市, 2019. Accessed on: Jul. 15, 2021. [Online]. Available: https://hdl.handle.net/11296/6484zw
    [36] E. D, Unity vs Unreal : 兩大遊戲引擎優缺點比較, Marketing & Design KM, Dec. 24, 2020. Accessed on: Jul. 14, 2021. [Online]. Available: http://13.231.129.69/2020/12/24/unity-vs-unreal-%E5%85%A9%E5%A4%A7%E9%81%8A%E6%88%B2%E5%BC%95%E6%93%8E%E5%84%AA%E7%BC%BA%E9%BB%9E%E6%AF%94%E8%BC%83/
    [37] ePrice, Samsung Galaxy S6 64G 價格、評價、規格 | ePrice 比價王, ePrice, May. 19, 2016. Accessed on: Jul. 18, 2021. [Online]. Available: https://www.eprice.com.tw/mobile/intro/c01-p5213-samsung-galaxy-s6-64g/
    [38] ePrice, ASUS ZenFone 4 (2017, ZE554KL) 4GB/64GB 價格、評價、規格 | ePrice 比價王, ePrice, May. 2, 2019. Accessed on: Jul. 18, 2021. [Online]. Available: https://www.eprice.com.tw/mobile/intro/c01-p5725-asus-zenfone-4-2017-ze554kl-4gb_64gb/
    [39] WangHenry娛樂數位, 【Gear VR-開箱】行動版虛擬實境(Virtual Reality),『Samsung Gear VR』遲來的開箱文(圖)!, WangHenry娛樂數位, May. 16, 2016. Accessed on: Jul. 18, 2021. [Online]. Available: https://www.wanghenry.com/2016/05/virtual-reality-samsung-gear-vr.html
    [40] Google, 取得 Cardboard – Google VR, Google, Jun. 16, 2021. Accessed on: Jul. 18, 2021. [Online]. Available: https://arvr.google.com/intl/zh-TW_tw/cardboard/get-cardboard/
    [41] 手機王SOGI, SAMSUNG Gear VR, 手機王SOGI, Aug. 5, 2021. Accessed on: Aug. 5, 2021. [Online]. Available: https://www.sogi.com.tw/products/samsung_gear_vr/11889
    [42] 手機王SOGI, Google Cardboard, 手機王SOGI, Aug. 5, 2021. Accessed on: Aug. 5, 2021. [Online]. Available: https://www.sogi.com.tw/products/google_cardboard/12179
    [43] 王智偉, FPS幀數是什麼?24fps、30fps、60fps有什麼區別?, Rene.E Laboratory, Jun. 4, 2019. Accessed on: Jul. 18, 2021. [Online]. Available: https://www.reneelab.net/frames-per-second.html#Ab
    [45] 手機王SOGI, SAMSUNG GALAXY S6 64GB, 手機王SOGI, Jul. 18, 2021. Accessed on: Jul. 18, 2021. [Online]. Available: https://www.sogi.com.tw/products/samsung_galaxy_s6_64gb/11438
    [47] ASUS, Zenfone 8 - 產品規格|智慧手機|ASUS 台灣, ASUS, Aug. 4, 2021. Accessed on: Aug. 5, 2021. [Online]. Available: https://www.asus.com/tw/Mobile/Phones/ZenFone/Zenfone-8/techspec/
    [48] 艷后潔思米, ASUS Zenfone 8、Zenfone 8 Flip 開箱 規格比較 延續翻轉鏡頭滿足拍攝需求 - 我的智慧家庭在myfone購物|達人教你買3C, 我的智慧家庭在myfone購物, May. 25, 2021. Accessed on: Aug. 5, 2021. [Online]. Available: https://myfone.blog/asus-zenfone-8-series-5g-unbox/
    [49] 手機王SOGI, ASUS Zenfone 8 ZS590KS (8GB/128GB), 手機王SOGI, Aug. 5, 2021. Accessed on: Aug. 5, 2021. [Online]. Available: https://www.sogi.com.tw/products/asus_zenfone_8_zs590ks/15852

    英文參考文獻
    [3] B. Dickson, Tesla AI Chief Explains Why Self-driving Cars Don’t Need Lidar, TechTalks, Jun. 28, 2021. Accessed on: Jul. 13, 2021. [Online]. Available: https://bdtechtalks.com/2021/06/28/tesla-computer-vision-autonomous-driving/
    [9] N.-T. Do, S.-H. Kim, H.-J. Yang, and G.-S. Lee, "Robust Hand Shape Features for Dynamic Hand Gesture Recognition Using Multi-Level Feature LSTM," Applied Sciences, vol. 10, no. 18, Sep. 2020. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.3390/app10186293.
    [10] L. Song, R. Hu, Y. Xiao, and L. Gong, "Real-Time 3D Hand Gesture Recognition from Depth Image," Advanced Materials Research, vol. 765-767, Apr. 2013. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.2991/icsem.2013.242.
    [12] T. Simon, H. Joo, I. Matthews, and Y. Sheikh, "Hand Keypoint Detection in Single Images Using Multiview Bootstrapping," in Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, Honolulu, 2017, pp. 1145-1153. Accessed on: Jul. 13, 2021. [Online]. Available: http://openaccess.thecvf.com/content_cvpr_2017/html/Simon_Hand_Keypoint_Detection_CVPR_2017_paper.html
    [13] Z. Cao, G. Hidalgo, T. Simon, S. E. Wei, and Y. Sheikh, "OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 172-186, Jan. 2021. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.1109/TPAMI.2019.2929257.
    [14] F. Zhang, V. Bazarevsky, A. Vakunov, A. Tkachenka, G. Sung, C.-L. Chang, and M. Grundmann, "MediaPipe Hands: On-device Real-time Hand Tracking," arXiv e-prints, p. arXiv:2006.10214, Jun. 2020. Accessed on: Jul. 13, 2021. [Online]. Available: https://ui.adsabs.harvard.edu/abs/2020arXiv200610214Z.
    [15] S. Yousefi, M. Kidane, Y. Delgado, J. Chana, and N. Reski, "3D Gesture-based Interaction for Immersive Experience in Mobile VR," in 2016 23rd International Conference on Pattern Recognition (ICPR), Cancún, 2016, pp. 2121-2126. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.1109/ICPR.2016.7899949
    [16] A. Juliani, V.-P. Berges, E. Teng, A. Cohen, J. Harper, C. Elion, C. Goy, Y. Gao, H. Henry, and M. Mattar, "Unity: A General Platform for Intelligent Agents," arXiv preprint arXiv:1809.02627, Sep. 2018. Accessed on: Sep. 6, 2021. [Online]. Available: https://ui.adsabs.harvard.edu/abs/2018arXiv180902627J.
    [17] A. Mehrfard, J. Fotouhi, G. Taylor, T. Forster, N. Navab, and B. Fuerst, "A Comparative Analysis of Virtual Reality Head-Mounted Display Systems," arXiv e-prints, p. arXiv:1912.02913, Dec. 2019. Accessed on: Jul. 13, 2021. [Online]. Available: https://ui.adsabs.harvard.edu/abs/2019arXiv191202913M.
    [18] N. M. Papachristos, I. Vrellis, and T. A. Mikropoulos, "A Comparison between Oculus Rift and a Low-Cost Smartphone VR Headset: Immersive User Experience and Learning," in 2017 IEEE 17th International Conference on Advanced Learning Technologies (ICALT), Timisoara, 2017, pp. 477-481. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.1109/ICALT.2017.145
    [20] Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, and L. D. Jackel, "Backpropagation Applied to Handwritten Zip Code Recognition," Neural Computation, vol. 1, no. 4, pp. 541-551, Dec. 1989. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.1162/neco.1989.1.4.541.
    [21] A. Ouaknine, Review of Deep Learning Algorithms for Object Detection, Feb. 5, 2018. Accessed on: Jul. 13, 2021. [Online]. Available: https://medium.com/zylapp/review-of-deep-learning-algorithms-for-object-detection-c1f3d437b8522
    [22] L. Wang, Support vector machines: theory and applications. Berlin: Springer Science & Business Media, 2005.
    [23] Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based Learning Applied to Document Recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, Nov. 1998. Accessed on: Jul. 13, 2021. [Online]. Available: doi: 10.1109/5.726791.
    [24] A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet Classification with Deep Convolutional Neural Networks," Advances in neural information processing systems, vol. 25, pp. 1097-1105, 2012.
    [25] K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," arXiv e-prints, p. arXiv:1409.1556, Sep. 2014. Accessed on: Jul. 13, 2021. [Online]. Available: https://ui.adsabs.harvard.edu/abs/2014arXiv1409.1556S.
    [26] T. Chansanchai, Windows 10 Anniversary Update, NASCAR’s new race app, LGBT pride — Weekend Reading: July 1 edition, Microsoft, Jul. 1, 2016. Accessed on: Jul. 13, 2021. [Online]. Available: https://blogs.microsoft.com/blog/2016/07/01/windows-10-anniversary-update-nascars-new-race-app-lgbt-pride-weekend-reading-july-1-edition/
    [27] V. Mühler, Simple Hand Gesture Recognition using OpenCV and JavaScript., Sep. 5, 2017. Accessed on: Jul. 14, 2021. [Online]. Available: https://medium.com/@muehler.v/simple-hand-gesture-recognition-using-opencv-and-javascript-eb3d6ced28a0
    [28] A. Fitzgibbon, E. Krupka, J. Shotton, T. Cashman, T. Sharp, and D. Sweeney, Fully Articulated Hand Tracking, Microsoft, Oct. 2, 2014. Accessed on: Jul. 14, 2021. [Online]. Available: https://www.microsoft.com/en-us/research/project/fully-articulated-hand-tracking/
    [29] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You Only Look Once: Unified, Real-time Object detection," in Proceedings of the IEEE conference on computer vision and pattern recognition, Las Vegas, 2016, pp. 779-788. Accessed on: Jul. 13, 2021. [Online]. Available: https://www.cv-foundation.org/openaccess/content_cvpr_2016/html/Redmon_You_Only_Look_CVPR_2016_paper.html
    [30] J. Nelson and J. Solawetz, YOLOv5 is Here: State-of-the-Art Object Detection at 140 FPS, Roboflow, Jun. 10, 2020. Accessed on: Jul. 13, 2021. [Online]. Available: https://blog.roboflow.com/yolov5-is-here/
    [31] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, and A. C. Berg, "Ssd: Single Shot Multibox Detector," in European conference on computer vision, Amsterdam, 2016, pp. 21-37. Accessed on: Jul. 13, 2021. [Online]. Available: https://link.springer.com/chapter/10.1007/978-3-319-46448-0_2
    [32] MediaPipe, MediaPipe Hands, MediaPipe, Jun. 29, 2021. Accessed on: Jul. 13, 2021. [Online]. Available: https://google.github.io/mediapipe/solutions/hands
    [35] J. Haas, "A History of the Unity Game Engine," Ph.D. dissertation, Worcester Polytechnic Institute, Worcester, 2014. Accessed on: Jul. 15, 2021. [Online]. Available: https://web.wpi.edu/Pubs/E-project/Available/E-project-030614-143124/unrestricted/Haas_IQP_Final.pdf
    [44] L. E. Humes, T. A. Busey, J. C. Craig, and D. Kewley-Port, "The Effects of Age on Sensory Thresholds and Temporal Gap Detection in Hearing, Vision, and Touch," Attention, Perception, & Psychophysics, vol. 71, no. 4, pp. 860-871, May. 2009. Accessed on: Jul. 18, 2021. [Online]. Available: https://link.springer.com/article/10.3758/APP.71.4.860.
    [46] Qualcomm, Snapdragon 630 Mobile Platform, Qualcomm, Jul. 18, 2021. Accessed on: Jul. 18, 2021. [Online]. Available: https://www.qualcomm.com/products/snapdragon-630-mobile-platform

    無法下載圖示 Full text public date 2024/09/19 (Intranet public)
    Full text public date This full text is not authorized to be published. (Internet public)
    Full text public date This full text is not authorized to be published. (National library)
    QR CODE