The File
Metadata
Creators:
- Perfler, Felix
, Acoustics Research Institute
- Pausch, Florian
, Acoustics Research Institute
- Holighaus, Nicki
, Acoustics Research Institute
- Majdak, Piotr
, Acoustics Research Institute
Publishers:
Rightsholders:
Keywords:
- Physically based modeling
- Computer vision
- Neural network
- Vision Transformer
- Supervised learning
- Regression
Relations:
- Mesh2PPM 1.0.0 is derived from https://github.com/Any2HRTF/Mesh2PPM (URL).
Other:
- DOI: not assigned yet
- Uploaded by: Florian Pausch
- Date (created): 2025-06-29 21:55:14 (GMT)
- Date (updated): 2025-12-02 09:38:22 (GMT)
- Production Year: 2022-2024
- Resource Type: Model
- Rights: EUPL-1.2: European Union Public Licence version 1.2
- Subject Areas: Life Science , Other SONICOM Ecosystem
- General Description: Tool Mesh2PPM 1.0.0 takes synthetic pinna geometries generated with Tool PyBezierPPM 3.0 as input to predict the local parameters of Tool BezierPPM 3.0.
- Methods: Tool Mesh2PPM 1.0.0 consists of a rendering stage and a deep neural network (DNN). The synthetic pinna geometries at the input are rendered with Tool PyBezierPPM 3.0 as multi-view grey-scale pinna images and optionally as depth images, i.e., MVPD pinna images. The DNN uses these MVPD pinna images as input and predicts the local parameters of Tool BezierPPM 3.0. The Database Mesh2PPM 1.0.0: Weights provides weights for various DNN instances trained for specific experimental conditions.
- Technical Remarks: Implemented in PyTorch.
Comments
No comments found.
Uploaded by: Florian Pausch
Created: 2025-06-29 21:55:14
Updated: 2025-12-02 09:38:22