Articulated Vehicle Detection via Learning Inter-Object Relationships Using GNN in LiDAR Point Clouds
LiDAR点群におけるGNNを用いた物体間の関係性の学習による連接車両の検出
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20265390
- Paper/Info type
- Proceedings (Spring)
No.90-26
- Pages
- 1-6(Total 6 p)
- Date of publication
- May 2026
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2026 JSAE Annual Congress (Spring)
Detailed Information
| Author(J) | 1) 篭橋 陸, 2) 針屋 慶吾, 3) 米陀 佳祐, 4) 福田 有輝也, 5) 菅沼 直樹 |
|---|---|
| Author(E) | 1) Riku Kagohashi, 2) Keigo Hariya, 3) Keisuke Yoneda, 4) Yukiya Fukuda, 5) Naoki Suganuma |
| Affiliation(J) | 1) 金沢大学, 2) 金沢大学, 3) 金沢大学, 4) 金沢大学, 5) 金沢大学 |
| Affiliation(E) | 1) Kanazawa University, 2) Kanazawa University, 3) Kanazawa University, 4) Kanazawa University, 5) Kanazawa University |
| Abstract(J) | LiDAR点群における連接車両の検出は，形状の複雑さや自己遮蔽による情報の欠落に伴う検出精度低下が課題である．本研究では，Graph Neural Network（GNN) を用いて物体間の関係性を学習する物体検出モデルを提案し，個々の形状情報に加え車両間の配置や文脈情報を統合的に考慮することで，連接車両の検出精度向上を目的とする． Translation |
| Abstract(E) | Detecting articulated vehicles in LiDAR point clouds suffers from low accuracy due to geometric complexity and information loss caused by self-occlusion. This study proposes an object detection model that learns inter-object relationships using Graph Neural Networks (GNN). By jointly considering spatial arrangement and contextual information in addition to individual geometric features, we aim to improve the detection accuracy of articulated vehicles. |
