Deformasi Model Wajah Berbasis Landmark Menggunakan Laplacian

Aji, Ongki Permono (2019) Deformasi Model Wajah Berbasis Landmark Menggunakan Laplacian. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 07111650052002-Master_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Merekonstruksi model wajah 3D dari tengkorak yang tidak dikenal memberikan banyak manfaat di bidang arkeologi, antropologi, dan investigasi forensik. Sistem komputer untuk rekonstruksi wajah 3D memberikan keuntungan besar dalam mengurangi konsumsi waktu. Tujuan dari tesis ini adalah untuk memprediksi bentuk wajah 3D orang Indonesia berbasis komputer dari gambar CT Scan menggunakan deformasi model wajah yang dibatasi oleh pengetahuan tentang ketebalan jaringan lunak. Data tengkorak kepala dan wajah diperoleh dari hasil pemindaian menggunakan CT Scan. Kemudian diambil struktur tengkorak dan wajahnya dan disimpan dalam bentuk point cloud. Pertama, kami mempersiapkan data pendukung berupa model wajah umum untuk orang Indonesia dan FSTT orang Indonesia. Kemudian kami melakukan penandaan landmark terhadap tengkorak yang tidak dikenal. Dengan menambahkan FSTT terhadap landmark tengkorak tersebut, kami mendapatkan koordinat landmark dari prediksi wajah. Terakhir, kami melakukan deformasi Laplacian terhadap model wajah supaya bentuk dan ukurannya menyesuaikan dengan koordinat landmark dari prediksi wajah tersebut.
=================================================================================================================================
Reconstructing a 3D face model from an unknown skull provides many benefits in the fields of archeology, anthropology, and forensic investigation. Computer systems for 3D facial reconstruction provide a big advantage in reducing time consumption. The purpose of this thesis is to produce computer-based 3D faces predictions of Indonesians from CT scans using facial model deformation which is limited by knowledge of soft tissue thickness. Skull and face data were obtained from scanning using a CT scan. Then, the skull and face structure were taken and stored in the form of point cloud. First, we prepare supporting data in the form of general face models for Indonesians and Indonesian FSTTs. Then we set landmarks on unknown skulls. By adding FSTT to the skull landmarks, we get landmark coordinates from face predictions. Finally, we performed Laplacian deformation on the face model so that its shape and size matched with the landmark coordinates of the face's predictions.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	laplacian, surface deformation, point cloud, FSTT, landmark
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.B56 Biometric identification
Divisions:	Faculty of Electrical Technology > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Ongki Permono Aji
Date Deposited:	23 Jul 2026 08:09
Last Modified:	23 Jul 2026 08:09
URI:	http://repository.its.ac.id/id/eprint/66833

Actions (login required)

View Item