Đồ án tốt nghiệp - Rit3d: a real-time indoor 3d point cloud method for efficient object detection

HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION
FACULTY OF INTERNATIONAL EDUCATION

GRADUATION THESIS

RIT3D: A REAL-TIME INDOOR 3D POINT CLOUD METHOD
FOR EFFICIENT OBJECT DETECTION

COMPUTER ENGINEERING

Student: MAI THANH LAM
Student ID: 20119137
MAI HONG PHONG
Student ID: 20119192
Major: COMPUTER ENGINEERING

HO CHI MINH CITY, 07/2025

Advisor: Ph.D. HUYNH THE THIEN

HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION
FACULTY OF INTERNATIONAL EDUCATION

GRADUATION THESIS

RIT3D: A REAL-TIME INDOOR 3D POINT CLOUD METHOD
FOR EFFICIENT OBJECT DETECTION

COMPUTER ENGINEERING

Student: MAI THANH LAM
Student ID: 20119137
MAI HONG PHONG
Student ID: 20119192
Major: COMPUTER ENGINEERING

HO CHI MINH CITY, 07/2025

Advisor: Ph.D. HUYNH THE THIEN

ACKNOWLEDGE
First and foremost, we would like to express our deepest gratitude to our
advisor, Ph.D. Huynh The Thien, for his invaluable guidance, encouragement,
and constant support throughout the duration of this thesis. His expertise and
insights have been instrumental in shaping the direction and quality of our
research.
We are also thankful to the lecturers and staff of the Faculty of Electrical
Engineering and the Department of Electronics and Telecommunications at Ho
Chi Minh City University of Technology and Education for providing us with
a solid academic foundation and a supportive learning environment.
Special thanks go to our families and friends for their unwavering support,
understanding, and encouragement during the challenging periods of our study.
This thesis would not have been possible without the knowledge and
resources made available by the academic and research community in the field
of 3D computer vision.
STUDENT
(Sign and clearly state the full name)

Mai Thanh Lam Mai Hong Phong

DECLARATION
We hereby that this graduation thesis is the result of our own
independent work, conducted under the supervision of Ph.D. Huynh The Thien.
All sources of information and references used in this thesis have been
appropriately acknowledged. We confirm that the content of this thesis has
not been submitted for any other degree or qualification at this university or any
other institution.
We take full responsibility for the content and results presented in this
research.
STUDENT
(Sign and clearly state the full name)

Mai Thanh Lam Mai Hong Phong

ABSTRACT
3D object detection in point cloud data is a foundational task for indoor scene
understanding in applications
...

...
in, “Dense rgb-d semantic mapping with pixel-voxel neural
network,” in 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS),
2017, pp. 1–8.
[52] C. Choy, J. Gwak, and S. Savarese, “4d spatio-temporal convnets:
Minkowski
convolutional neural networks,”
in Proceedings of the IEEE Conference on
Computer Vision and Pattern Recognition (CVPR), 2019, pp. 3075–3084. [Online].
Available: https://openaccess.thecvf.com/content_CVPR_2019/html/Choy_4D_Spatio-Temporal_
ConvNets_Minkowski_Convolutional_Neural_Networks_CVPR_2019_paper.html
[53] B. Graham, M. Engelcke, and L. van der Maaten, “3d semantic segmentation with submanifold
sparse convolutional networks,” in Proceedings of the IEEE Conference on Computer Vision and
Pattern Recognition (CVPR), 2018, pp. 9224–9232. [Online]. Available: https://openaccess.thecvf.
com/content_cvpr_2018/html/Graham_3D_Semantic_Segmentation_CVPR_2018_paper.html
[54] H. Tang, Z. Liu, X. Li, Y. Lin, and S. Han, “Torchsparse: Efficient point cloud inference engine,”
in Proceedings of the 5th Conference on Machine Learning and Systems (MLSys), 2022, accessed:
2025-07-04. [Online]. Available: https://hanlab.mit.edu/projects/torchsparse
[55] Y. Yan, Y. Mao, and B. Li, “SECOND: Sparsely embedded convolutional detection,” Sensors,
vol. 18, no. 10, p. 3337, 2018. [Online]. Available: https://www.mdpi.com/1424-8220/18/10/3337
[56] H. Tang, S. Yang, Z. Liu, K. Hong, Z. Yu, X. Li, G. Dai, Y. Wang, and S. Han, “Torchsparse++:
Efficient training and inference framework for sparse convolution on gpus,” 2023. [Online].
Available: https://arxiv.org/abs/2311.12862
[57] Y. Lee, J.-W. Hwang, S. Lee, Y. Bae, and J. Park, “An energy and GPU-Computation efficient
backbone network for Real-Time object detection,” 2022 IEEE/CVF Conference on Computer
Vision and Pattern Recognition Workshops (CVPRW), pp. 752–760, 6 2019. [Online]. Available:
https://doi.org/10.1109/cvprw.2019.00103
[58] M. Contributors, “MMDetection3D: OpenMMLab next-generation platform for general 3D object
detection,” in GitHub Repository, 2020. [Online]. Available: https://github.com/open-mmlab/
mmdetection3d
[59] I. Loshchilov and F. Hutter, “Decoupled Weight Decay Regularization,” in International
Conference on Learning Representations, 2019. [Online]. Available: https://openreview.net/forum?
id=Bkg6RiCqY7
[60] J. Gwak, C. Choy, and S. Savarese, “Generative sparse detection networks for 3d single-shot object
detection,” in ECCV, 2020.

39