 |
International Journal for Research Trends and Innovation
International Peer Reviewed & Refereed Journals, Open Access Journal
ISSN Approved Journal No: 2456-3315 | Impact factor: 8.14 | ESTD Year: 2016
Scholarly open access journals, Peer-reviewed, and Refereed Journals, Impact factor 8.14 (Calculate by google scholar and Semantic Scholar | AI-Powered Research Tool) , Multidisciplinary, Monthly, Indexing in all major database & Metadata, Citation Generator, Digital Object Identifier(DOI)
|
Issue: December 2025
Volume 10 | Issue 12
Review Result and Publication of Paper within : 2-3 days
Click Here For more DetailsFor Authors
Forms / Download
Published Issue Details
Editorial Board
Other IMP Links
Facts & Figure
Impact Factor : 8.14
Issue per Year : 12
Volume Published : 10
Issue Published : 115
Article Submitted : 19455
Article Published : 8041
Total Authors : 21252
Total Reviewer : 769
Total Countries : 144
Indexing Partner
Licence
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
Published Paper Details
|
|
| Paper Title: | Dynamic Edge Weights and Density Clustering: Advancing Single-Stage 3D Object Detection with Graph Neural Networks in Point Clouds |
| Authors Name: | Akash Ghosh , Shreya Vats , Manmeet Singh , Shashank Patoju , Dheeraj Kumar Alla, Vinod Kumar |
| Download E-Certificate: | Download |
| Author Reg. ID: |
IJRTI_188686
|
| Published Paper Id: | IJRTI2312038 |
| Published In: | Volume 8 Issue 12, December-2023 |
| DOI: | |
| Abstract: | This paper presents a novel single-stage 3D object detection framework that addresses the underdeveloped landscape of point-based networks. Current approaches in the point cloud domain often overlook optimization, impeding the comprehensive exploration of relationships among adjacent point sets. The proposed method integrates density clustering and graph neural networks (GNNs) to form an efficient single-stage 3D object detection network. Employing a density clustering ball query optimizes the partitioned point cloud space, emphasizing the inclusion of key point sets harboring detailed object features. Leveraging GNNs facilitates the extraction of both local and global relationships inherent in the data. The approach showcases rapid inference speeds and achieves a superior balance between detection performance and inference time, as demonstrated through experiments on the KITTI dataset. Furthermore, our paper introduces Point-GNN, a graph neural network designed for object detection in LiDAR point clouds. The approach efficiently encodes the point cloud using a fixed-radius near-neighbors graph and leverages the proposed Point-GNN to predict object categories and shapes for each graph vertex. Auto-registration mechanisms reduce translation variance, and a box merging and scoring operation accurately combines detections from multiple vertices. Experimental results on the KITTI benchmark reveal that Point-GNN surpasses fusion-based algorithms, showcasing the potential of graph neural networks in the realm of 3D object detection. Additionally, the paper contributes to the field by proposing an attention-based feature aggregation technique within a graph neural network for 3D object detection in LiDAR scans. The method employs a distance-aware down-sampling scheme to enhance algorithmic performance while retaining maximum geometric features, even for objects distant from the sensor. Each GNN layer incorporates per-node masked attention, considering the underlying neighborhood graph structure and eliminating costly matrix operations. The proposed method achieves comparable results for 3D object detection, demonstrating its robustness and accuracy on the KITTI dataset. |
| Keywords: | KITTI dataset, Graph neural network, Density Clustering, 3D object detection, Geometric Features |
| Cite Article: | "Dynamic Edge Weights and Density Clustering: Advancing Single-Stage 3D Object Detection with Graph Neural Networks in Point Clouds", International Journal of Science & Engineering Development Research (www.ijrti.org), ISSN:2455-2631, Vol.8, Issue 12, page no.274 - 282, December-2023, Available :http://www.ijrti.org/papers/IJRTI2312038.pdf |
| Downloads: | 000205167 |
| ISSN: |
2456-3315 | IMPACT FACTOR: 8.14 Calculated By Google Scholar| ESTD YEAR: 2016 An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.14 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator |
| Publication Details: |
Published Paper ID: IJRTI2312038
Registration ID:188686
Published In: Volume 8 Issue 12, December-2023
DOI (Digital Object Identifier):
Page No: 274 - 282 Country: Delhi, Delhi, India Research Area: Science & Technology Publisher : IJ Publication Published Paper URL : https://www.ijrti.org/viewpaperforall?paper=IJRTI2312038 Published Paper PDF: https://www.ijrti.org/papers/IJRTI2312038 |
| Share Article: | |
|
Click Here to Download This Article |
|
| Article Preview | |
|
|
|
Major Indexing from www.ijrti.org
| Google Scholar | ResearcherID Thomson Reuters | Mendeley : reference manager | Academia.edu |
| arXiv.org : cornell university library | Research Gate | CiteSeerX | DOAJ : Directory of Open Access Journals |
| DRJI | Index Copernicus International | Scribd | DocStoc |
ISSN Details
 |
 |
|
ISSN: 2456-3315
Impact Factor: 8.14 and ISSN APPROVED,
Journal Starting Year (ESTD) : 2016
DOI (A digital object identifier)
| Providing A digital object identifier by DOI.ONE How to Get DOI? |
Conference
Open Access License Policy
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Important Details
Join RMS/Earn 300
WhatsApp
Click Here
Click Here
Indexing Partner |
|||
| Copyright © 2025 - All Rights Reserved - IJRTI | |||