A Texture Superpixel Approach to Semantic Material Classification for Acoustic Geometry Tagging

Colombo, Mattia and Dolhasz, Alan and Harvey, Carlo (2021) A Texture Superpixel Approach to Semantic Material Classification for Acoustic Geometry Tagging. In: ACM CHI 2021, May 8-13, Online Virtual Conference (originally Yokohama, Japan). (In Press)

[img] Text
chi21c-sub1895-i7.pdf - Accepted Version
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

The current state of audio rendering algorithms allows efficient sound propagation, reflecting realistic acoustic properties of real environments. Among factors affecting realism of acoustic simulations is the mapping between an environment’s geometry, and acoustic information of materials represented. We present a pipeline to infer material characteristics from their visual representations, providing an automated mapping. A trained image classifier estimates semantic material information from textured meshes mapping predicted labels to a database of measured frequency-dependent absorption coefficients; trained on a material image patches generated from superpixels, it produces inference from meshes, decomposing their unwrapped textures. The most frequent label from predicted texture patches determines the acoustic material assigned to the input mesh. We test the pipeline on a real environment, capturing a conference room and reconstructing its geometry from point cloud data. We estimate a Room Impulse Response (RIR) of the virtual environment, which we compare against a measured counterpart.

Item Type: Conference or Workshop Item (Paper)
Date: 18 February 2021
Subjects: G400 Computer Science
G700 Artificial Intelligence
Divisions: Faculty of Computing, Engineering and the Built Environment > School of Computing and Digital Technology > Digital Media Technology
Depositing User: Carlo Harvey
Date Deposited: 16 Mar 2021 10:36
Last Modified: 16 Mar 2021 10:37
URI: http://www.open-access.bcu.ac.uk/id/eprint/11281

Actions (login required)

View Item View Item

Research

In this section...