3D Point Cloud Network Pruning: When Some Weights Do not Matter


Amrijit Biswas (North South University), Md. Ismail Hossain (North South University), M M Lutfe Elahi (North South University), Ali Cheraghian (CSIRO), Fuad Rahman (University of Arizona), Nabeel Mohammed (North South University), Shafin Rahman (North South University)
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)

Abstract

A point cloud is a crucial geometric data structure utilized across numerous applications. The adoption of deep neural networks referred to as Point Cloud Neural Networks(PCNNs), for processing 3D point clouds has significantly advanced fields that rely on 3D geometric data to enhance the efficiency of tasks. Expanding the size of both neural network models and 3D point clouds introduces significant challenges in minimizing computational and memory requirements. This is essential for meeting the demanding requirements of real-world applications, which prioritize minimal energy consumption and low latency. Therefore, investigating redundancy in PCNNs is crucial yet challenging due to their sensitivity to parameters. Additionally, traditional pruning methods face difficulties as these networks rely heavily on weights and points. Nonetheless, our research reveals a promising phenomenon that could refine standard PCNN pruning techniques. Our findings suggest that preserving only the top p% of the highest magnitude weights is crucial for accuracy preservation. For example, pruning 99% of the weights from the PointNet model still results in accuracy close to the base level. Specifically, in the ModelNet40 dataset, where the base accuracy with the PointNet model was 87.5%, preserving just 1% of the weights still achieves an accuracy of 86.8%. Codes are available in: https://github.com/apurba-nsu-rnd-lab/PCNN_Pruning

Citation

@inproceedings{Biswas_2024_BMVC,
author    = {Amrijit Biswas and Md. Ismail Hossain and M M Lutfe Elahi and Ali Cheraghian and Fuad Rahman and Nabeel Mohammed and Shafin Rahman},
title     = {3D Point Cloud Network Pruning: When Some Weights Do not Matter},
booktitle = {35th British Machine Vision Conference 2024, {BMVC} 2024, Glasgow, UK, November 25-28, 2024},
publisher = {BMVA},
year      = {2024},
url       = {https://papers.bmvc2024.org/0637.pdf}
}


Copyright © 2024 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection