Research Article

Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation

Volume: 20 Number: 1 March 27, 2025
EN TR

Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation

Abstract

3D semantic segmentation, the process of assigning semantic labels to every point in a 3D space, is critical for numerous applications, including autonomous driving, robotics, medical imaging, and urban mapping. Despite significant progress, challenges such as data imbalance, scalability, and real-time processing constraints persist. This study addresses the real-time processing issue by comparing Tiny, Medium, and Large PointNet-inspired models utilizing the ShapeNetCore dataset. The models incorporate the T-Net module for pose normalization to maintain robustness against geometric transformations. Class-specific segmentation is explored by training separate models for the Airplane, Motorbike, and Car classes, allowing custom optimizations for each class. The Tiny model with 512 sampled points where the batch size is 16 and trained for 40 epochs with a starting learning rate of 1×10^(-3) achieved an average training accuracy of 86.18% and an average validation accuracy of 83.50%, making it optimal for real-time applications due to its fast inference speed and high accuracy.

Keywords

References

  1. Qi CR, Su H, Mo K, Guibas LJ. PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. Proc IEEE Conf Comput Vis Pattern Recognit (CVPR), 2017; Honolulu, HI, USA. 652-660.
  2. Qi CR, Su H, Yi L, Guibas LJ. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. Adv Neural Inf Process Syst (NIPS), 2017; Long Beach, CA, USA. 30.
  3. Choy C, Gwak JY, Savarese S. 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. Proc IEEE/CVF Conf Comput Vis Pattern Recognit (CVPR), 2019; Long Beach, CA, USA. 3075-3084.
  4. Behley J, Garbade M, Milioto A, Quenzel J, Behnke S, Stachniss C, Gall J. SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences. Proc IEEE/CVF Conf Comput Vis Pattern Recognit (CVPR), 2019; Long Beach, CA, USA. 9297-9307.
  5. Zhang Y, Zhou Z, David P, Yue X, Xi Z, Gong B, Foroosh H. PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation. Proc. IEEE/CVF Conf Comput Vis Pattern Recognit (CVPR), 2020; Virtual. 9601-9610.
  6. Cortinhal T, Tzelepis G, Aksoy EE. SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving Int Symp Vis Comput., 2020; San Diego, CA, USA. 207-222.
  7. Tang H, Liu Z, Zhao S, Lin Y, Lin J, Wang H, Han S. Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution Eur Conf Comput Vis (ECCV), 2020; Glasgow, UK. 685-702.
  8. Li S, Zhang C, He X. Shape-aware Semi-supervised 3D Semantic Segmentation for Medical Images. Med Image Comput Comput Assist Interv (MICCAI), 2020; Lima, Peru. 552–561.

Details

Primary Language

English

Subjects

Image Processing

Journal Section

Research Article

Publication Date

March 27, 2025

Submission Date

February 11, 2025

Acceptance Date

March 18, 2025

Published in Issue

Year 2025 Volume: 20 Number: 1

APA
Karaman, F., & Gümüş, F. (2025). Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation. Turkish Journal of Science and Technology, 20(1), 327-337. https://doi.org/10.55525/tjst.1637713
AMA
1.Karaman F, Gümüş F. Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation. TJST. 2025;20(1):327-337. doi:10.55525/tjst.1637713
Chicago
Karaman, Furkan, and Fatma Gümüş. 2025. “Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation”. Turkish Journal of Science and Technology 20 (1): 327-37. https://doi.org/10.55525/tjst.1637713.
EndNote
Karaman F, Gümüş F (March 1, 2025) Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation. Turkish Journal of Science and Technology 20 1 327–337.
IEEE
[1]F. Karaman and F. Gümüş, “Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation”, TJST, vol. 20, no. 1, pp. 327–337, Mar. 2025, doi: 10.55525/tjst.1637713.
ISNAD
Karaman, Furkan - Gümüş, Fatma. “Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation”. Turkish Journal of Science and Technology 20/1 (March 1, 2025): 327-337. https://doi.org/10.55525/tjst.1637713.
JAMA
1.Karaman F, Gümüş F. Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation. TJST. 2025;20:327–337.
MLA
Karaman, Furkan, and Fatma Gümüş. “Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation”. Turkish Journal of Science and Technology, vol. 20, no. 1, Mar. 2025, pp. 327-3, doi:10.55525/tjst.1637713.
Vancouver
1.Furkan Karaman, Fatma Gümüş. Exploring the Impact of Model Capacity and Parameter Tuning on 3D Semantic Segmentation. TJST. 2025 Mar. 1;20(1):327-3. doi:10.55525/tjst.1637713