LOD-Net: Locality-Aware 3D Object Detection Using Multi-Scale Transformer Network
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911603604062208 |
|---|---|
| author | Khan, Mustaqeem Nurakhmetova, Aidana Gueaieb, Wail Saddik, Abdulmotaleb El |
| author_facet | Khan, Mustaqeem Nurakhmetova, Aidana Gueaieb, Wail Saddik, Abdulmotaleb El |
| contents | 3D object detection in point cloud data remains a challenging task due to the sparsity and lack of global structure inherent in the input. In this work, we propose a novel Multi-Scale Attention (MSA) mechanism integrated into the 3DETR architecture to better capture both local geometry and global context. Our method introduces an upsampling operation that generates high-resolution feature maps, enabling the network to better detect smaller and semantically related objects. Experiments conducted on the ScanNetv2 dataset demonstrate that our 3DETR + MSA model improves detection performance, achieving a gain of almost 1% in mAP@25 and 4.78% in mAP@50 over the baseline. While applying MSA to the 3DETR-m variant shows limited improvement, our analysis reveals the importance of adapting the upsampling strategy for lightweight models. These results highlight the effectiveness of combining hierarchical feature extraction with attention mechanisms in enhancing 3D scene understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_16696 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LOD-Net: Locality-Aware 3D Object Detection Using Multi-Scale Transformer Network Khan, Mustaqeem Nurakhmetova, Aidana Gueaieb, Wail Saddik, Abdulmotaleb El Computer Vision and Pattern Recognition Artificial Intelligence Image and Video Processing 3D object detection in point cloud data remains a challenging task due to the sparsity and lack of global structure inherent in the input. In this work, we propose a novel Multi-Scale Attention (MSA) mechanism integrated into the 3DETR architecture to better capture both local geometry and global context. Our method introduces an upsampling operation that generates high-resolution feature maps, enabling the network to better detect smaller and semantically related objects. Experiments conducted on the ScanNetv2 dataset demonstrate that our 3DETR + MSA model improves detection performance, achieving a gain of almost 1% in mAP@25 and 4.78% in mAP@50 over the baseline. While applying MSA to the 3DETR-m variant shows limited improvement, our analysis reveals the importance of adapting the upsampling strategy for lightweight models. These results highlight the effectiveness of combining hierarchical feature extraction with attention mechanisms in enhancing 3D scene understanding. |
| title | LOD-Net: Locality-Aware 3D Object Detection Using Multi-Scale Transformer Network |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Image and Video Processing |
| url | https://arxiv.org/abs/2604.16696 |