Prototipe aplikasi ruang ganti pakaian virtual berbasis kinect = Virtual dressing room prototype application based on kinect

Rosiana, Rosiana (2014) Prototipe aplikasi ruang ganti pakaian virtual berbasis kinect = Virtual dressing room prototype application based on kinect. Bachelor thesis, Universitas Pelita Harapan.

	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (6MB)
	Text (Abstract) Abstract.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (611kB)
	Text (ToC) ToC.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (372kB)
	Text (Chapter1) Chapter1.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (287kB)
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (786kB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (883kB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (278kB)
	Text (Bibliography) Bibliography.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (320kB)
	Text (Publication-Agreement) Publication-Agreement.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (7MB)

Abstract

Aplikasi ruang ganti pakaian virtual merupakan sebuah alternatif untuk ruang ganti pakaian konvensional yang dianggap lebih praktis serta menghemat waktu. Diantara ketersediaan teknologi sensor baru-baru ini, sensor Kinect dengan kemampuan melacak sendi pengguna menarik untuk dikaji. Pertanyaan penelitian di sini adalah apakah data kerangka (skeleton) hasil pendeteksian Kinect dapat digunakan sebagai koordinat acuan penempatan pakaian virtual pada tubuh pengguna. Sebagai pendekatan untuk mengkaji akurasi peletakkan pakaian virtual pada tubuh pengguna dilibatkan hasil kalkulasi jarak dan tinggi pengguna oleh Kinect. Selisih tinggi pengguna yang dihitung Kinect dengan tinggi model mesh pakaian virtual yang sudah ada akan digunakan sebagai pengali dengan jarak pengguna untuk menyesuaikan ukuran pakaian virtual pada tubuh pengguna. Pengukuran tubuh lainnya adalah panjang badan, lebar dada dan lebar pinggang. Dari pengujian fungsional yang dilakukan, hasil menunjukkan bahwa pemasangan pakaian virtual pada tubuh pengguna cukup cocok. Pada pengujian jarak, Kinect dapat mengidentifikasi pengguna yang berdiri setidaknya 2,20 meter dan tidak lebih dari 3,60 meter. Secara keseluruhan, akurasi hasil pengukuran tubuh dengan Kinect adalah 99,49% ±0,38% untuk tinggi tubuh, 97,66% ±1,56% untuk panjang badan, 89,43% ±6,11% untuk lebar dada dan 92,14% ±3,36% untuk lebar pinggang. / A virtual dressing room application is an alternative to conventional clothes changing method which is considered more practical, as well as time saving. On the availability of recent sensor technology, it is interesting to explore the functionality of Kinect whose sensors are known of being capable for tracking human joints. The research question here is whether the joints data detected by Kinect are appropriate to serve as coordinate reference of the virtual clothing on any user’s body. As an approach to assess the placing accuracy of the virtual clothes to any user’s body, the calculations of the user’s distance from the Kinect and his or her height are used. The difference between the user’s height calculated by Kinect and the height of the existing mesh model will be used as a multiplier to the user’s distance so that the size of the virtual clothing for the particular user can be determined. The other body measurements involve torso, chest and waist. Through a series of functional tests, the result shows that the virtual clothes fitting is quite appropriate. On the distance measurement, Kinect is able to identify the user standing at least 2.20 meters away from it and no longer than 3.60 meters. Overall, the body measurements using Kinect on height, torso, chest and waist yields 99.49% ± 0.38%, 97.66% ± 1.56%, 89.43% ± 6.11% and 92.14% ± 3.36% of accuracy respectively.

Actions (login required)

View Item