arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.16100cs.CV
收藏

利用稀疏点-体素卷积搜索高效的3D结构

Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution

IIIS, Tsinghua University(清华大学交叉信息研究院);Massachusetts Institute of Technology(麻省理工学院)

Haotian Tang, Zhijian Liu, Shengyu Zhao, Yujun Lin, Ji Lin, Hanrui Wang, Song Han

摘要:

自动驾驶汽车需要高效准确地理解3D场景,以便安全驾驶。给定有限的硬件资源,现有的3D感知模型不能识别小的实例(例如,行人,骑自行车的人)非常好,由于低分辨率体素化和积极的下采样。为此,我们提出了稀疏点体素卷积(SPVConv),这是一个轻量级的3D模块,它为香草稀疏卷积配备了高分辨率的基于点的分支。这种基于点的分支的开销可以忽略不计,即使是大型户外场景也能够保留精细的细节。为了探索高效的3D模型谱,我们首先基于SPVConv定义了一个灵活的架构设计空间,然后我们提出了3D神经架构搜索(3D-NAS),以在这个多样化的设计空间中高效地搜索最佳网络架构。实验结果验证了所得到的SPVNAS模型是快速和准确的:它比最先进的MinkowskiNet高出3.3%,在竞争激烈的SemanticKITTI排行榜上排名第一。它还实现了比MinkowskiNet更高精度的8倍计算减少和3倍测量加速。最后,我们将我们的方法转移到3D目标检测,它实现了一致的改进,在KITTI的一个阶段的检测基线。

英文摘要:

Self-driving cars need to understand 3D scenes efficiently and accurately in order to drive safely. Given the limited hardware resources, existing 3D perception models are not able to recognize small instances (e.g., pedestrians, cyclists) very well due to the low-resolution voxelization and aggressive downsampling. To this end, we propose Sparse Point-Voxel Convolution (SPVConv), a lightweight 3D module that equips the vanilla Sparse Convolution with the high-resolution point-based branch. With negligible overhead, this point-based branch is able to preserve the fine details even from large outdoor scenes. To explore the spectrum of efficient 3D models, we first define a flexible architecture design space based on SPVConv, and we then present 3D Neural Architecture Search (3D-NAS) to search the optimal network architecture over this diverse design space efficiently and effectively. Experimental results validate that the resulting SPVNAS model is fast and accurate: it outperforms the state-of-the-art MinkowskiNet by 3.3%, ranking 1st on the competitive SemanticKITTI leaderboard. It also achieves 8x computation reduction and 3x measured speedup over MinkowskiNet with higher accuracy. Finally, we transfer our method to 3D object detection, and it achieves consistent improvements over the one-stage detection baseline on KITTI.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑