Izračun HRTF za simulacijo prostorskega zvoka na podlagi zajema slik glave

Jovanović, Dušan (author), Klemenc, Bojan (mentor)

V diplomski nalogi je razvit sistem, ki iz fotografij, zajetih z mobilno aplikacijo, rekonstruira 3D-model glave in na njegovi geometriji izračuna individualiziran HRTF (nabor transformacij, ki opisujejo, kako se zvok spremeni zaradi oblike poslušalčeve glave in uhljev, preden pride do srednjega ušesa). Odjemalec na telefonu zajame in prek TCP pošlje vsaj 40 slik, strežnik pa izvede predobdelavo, maskiranje, rekonstrukcijo s COLMAP (SfM/MVS oz. rekonstrukcija strukture iz gibanja in stereo-rekonstrukcija z več pogledov), čiščenje in glajenje mreže ter izračun z orodjem Mesh2HRTF. Rezultati se shranijo v formatu SOFA. Kvalitativno testiranje z vizualizacijami HRIR (impulzni odziv glede na položaj glave) potrdi prisotnost ITD/ILD (medušesne časovne in nivojske razlike) in pokaže občutljivost na kakovost mreže.

• Language: Slovenian
• Keywords: HRTF, HRIR, 3D-rekonstrukcija, COLMAP, Mesh2HRTF, SOFA, binauralni zvok, računalniški vid, mobilna aplikacija, odjemalec-strežnik
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2026
• PID: 20.500.12556/RUL-180303
• COBISS.SI-ID: 271753987
• Publication date in RUL: 05.03.2026

Secondary language

• Language: English
• Title: Calculation of HRTF for simulating spatial audio based on head image acquisition

Abstract

This thesis presents a system that reconstructs a 3D head model from photographs captured with a mobile application and computes an individualized HRTF (head related transfer function) from the resulting geometry. The client application captures and sends at least 40 images via TCP, While the server performs preprocessing, masking, COLMAP-based reconstruction (SfM/MVS, Structure-from-Motion/Multi-View Stereo), mesh cleaning and smoothing, and finally runs the calculation using Mesh2HRTF. The output is stored in the SOFA format. Qualitative testing with HRIR visualizations confirms the presence of ITD/ILD (Interaural Time Difference / Interaural Level Difference) cues and highlights the strong dependence of the result on mesh quality.

• Keywords: HRTF, HRIR, 3D reconstruction, COLMAP, Mesh2HRTF, SOFA, binaural audio, computer vision, mobile application, client--server