Point-based Acoustic Scattering for Interactive Sound Propagation via Surface Encoding

Authors: Hsien-Yu Meng, Zhenyu Tang, Dinesh Manocha

IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We have analyzed the accuracy and perform validation on thousands of unseen 3D objects and highlight the benefits over other point-based geometric deep learning methods. ... We have evaluated the performance on thousands of objects that are very different from the training database (with varying convexity, genus, shape, size, orientation) and observe high accuracy. We also perform an ablation study to highlight the benefits of our approach.
Researcher Affiliation Academia University of Maryland, College Park {mengxy19,zhy,dmanocha}@umd.edu
Pseudocode No No explicit pseudocode or algorithm blocks were found.
Open Source Code No No concrete access to source code (specific repository link or explicit code release statement) for the methodology described in this paper was found.
Open Datasets Yes We sample 100,000 3D objects from the ABC Dataset [Koch et al., 2019].
Dataset Splits Yes We split our dataset into training, validation and test set following a 8:1:1 ratio.
Hardware Specification Yes The additional runtime overhead of estimating the scattering field from neural networks is less than 1ms per object on a NVIDIA Ge Force RTX 2080 Ti GPU." and "Our neural network is trained on an NVIDIA Ge Force RTX 2080 Ti GPU
Software Dependencies No The paper mentions 'Fast BEM Acoustics' and 'pyshtools' (with a URL), but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes The Adam optimizer is used and it decays exponentially with decay rate and decay step equal to 0.9 and 10 #training examples, respectively.