研究人员开发了DeGuNet，这是一种新的超紧凑图像骨干网络，旨在提高自动驾驶系统中三维目标检测的效率和准确性。该即插即用模块通过结合稀疏感知特征提取，更有效地整合激光雷达和相机数据，该方法将多视图图像与激光雷达深度数据对齐，同时防止无效区域污染。在nuScenes数据集上的实验表明，DeGuNet可将GPU内存使用量显著降低高达66.5%，并将推理速度提高1.16倍，同时将检测准确性提高高达6.20个绝对mAP。 AI
影响 通过优化多模态数据融合，提高自动驾驶感知系统的效率和准确性。
排序理由 详细介绍新技术方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
LiDAR-based collaborative 3D perception in Vehicle-to-Everything (V2X) systems typically relies on fusing bird's-eye-view (BEV) features across agents. However, current BEV representations, typically extracted by LiDAR backbones trained from scratch, are geometry-dominated and la…
In autonomous driving perception, the fusion of LiDAR and camera modalities has become the dominant paradigm for 3D object detection. However, current multi-modal frameworks heavily rely on massive visual backbones pretrained on 2D semantic tasks. This reliance introduces substan…