FGU3R: Fine-Grained Fusion via Unified 3D Representation for Multimodal 3D Object Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Guoxin, Song, Ziying, Liu, Lin, Ou, Zhonghong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929665194590208
author Zhang, Guoxin
Song, Ziying
Liu, Lin
Ou, Zhonghong
author_facet Zhang, Guoxin
Song, Ziying
Liu, Lin
Ou, Zhonghong
contents Multimodal 3D object detection has garnered considerable interest in autonomous driving. However, multimodal detectors suffer from dimension mismatches that derive from fusing 3D points with 2D pixels coarsely, which leads to sub-optimal fusion performance. In this paper, we propose a multimodal framework FGU3R to tackle the issue mentioned above via unified 3D representation and fine-grained fusion, which consists of two important components. First, we propose an efficient feature extractor for raw and pseudo points, termed Pseudo-Raw Convolution (PRConv), which modulates multimodal features synchronously and aggregates the features from different types of points on key points based on multimodal interaction. Second, a Cross-Attention Adaptive Fusion (CAAF) is designed to fuse homogeneous 3D RoI (Region of Interest) features adaptively via a cross-attention variant in a fine-grained manner. Together they make fine-grained fusion on unified 3D representation. The experiments conducted on the KITTI and nuScenes show the effectiveness of our proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2501_04373
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FGU3R: Fine-Grained Fusion via Unified 3D Representation for Multimodal 3D Object Detection
Zhang, Guoxin
Song, Ziying
Liu, Lin
Ou, Zhonghong
Computer Vision and Pattern Recognition
Multimodal 3D object detection has garnered considerable interest in autonomous driving. However, multimodal detectors suffer from dimension mismatches that derive from fusing 3D points with 2D pixels coarsely, which leads to sub-optimal fusion performance. In this paper, we propose a multimodal framework FGU3R to tackle the issue mentioned above via unified 3D representation and fine-grained fusion, which consists of two important components. First, we propose an efficient feature extractor for raw and pseudo points, termed Pseudo-Raw Convolution (PRConv), which modulates multimodal features synchronously and aggregates the features from different types of points on key points based on multimodal interaction. Second, a Cross-Attention Adaptive Fusion (CAAF) is designed to fuse homogeneous 3D RoI (Region of Interest) features adaptively via a cross-attention variant in a fine-grained manner. Together they make fine-grained fusion on unified 3D representation. The experiments conducted on the KITTI and nuScenes show the effectiveness of our proposed method.
title FGU3R: Fine-Grained Fusion via Unified 3D Representation for Multimodal 3D Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.04373