arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.10210cs.CVcs.LG
收藏

基于点云的目标重识别

Object Re-Identification from Point Clouds

University of Waterloo(滑铁卢大学)

Benjamin Thérien, Chengjie Huang, Adrian Chow, Krzysztof Czarnecki

AI总结:

本文首次大规模研究基于点云的目标重识别,构建了成对图像与LiDAR数据集,并提出轻量级匹配头以构建跨模态可比较的重识别网络,在实时条件下达到接近图像重识别的性能。

摘要:

基于图像的目标重识别(ReID)在图像检索(监控、零售分析等)和多目标跟踪(自动驾驶、机器人等)等应用领域中发挥着关键作用。然而,额外或专门通过深度传感器感知世界的系统正变得越来越普遍,却缺乏相应的目标重识别方法。在这项工作中,我们填补了这一空白,首次对基于点云的目标重识别进行了大规模研究,并确立了其相对于图像重识别的性能。为了实现这一研究,我们创建了两个具有成对图像和LiDAR观测的大规模重识别数据集,并提出了一种轻量级匹配头,可以连接到任何集合或序列处理骨干网络(例如PointNet或ViT),从而为两种模态创建了一系列可比较的目标重识别网络。以孪生方式运行时,我们提出的点云重识别网络可以实时(10 Hz)进行数千次成对比较。我们的研究结果表明,其性能随着传感器分辨率的提高而提升,并且在观测足够密集时接近图像重识别的性能。我们在最大规模下训练的最强网络对刚性目标实现了超过90%的重识别准确率,对可变形目标(无需任何显式骨架归一化)实现了超过85%的准确率。据我们所知,我们是首个研究基于真实点云观测的目标重识别的团队。

英文摘要:

Object re-identification (ReID) from images plays a critical role in application domains of image retrieval (surveillance, retail analytics, etc.) and multi-object tracking (autonomous driving, robotics, etc.). However, systems that additionally or exclusively perceive the world from depth sensors are becoming more commonplace without any corresponding methods for object ReID. In this work, we fill the gap by providing the first large-scale study of object ReID from point clouds and establishing its performance relative to image ReID. To enable such a study, we create two large-scale ReID datasets with paired image and LiDAR observations and propose a lightweight matching head that can be concatenated to any set or sequence processing backbone (e.g., PointNet or ViT), creating a family of comparable object ReID networks for both modalities. Run in Siamese style, our proposed point cloud ReID networks can make thousands of pairwise comparisons in real-time ($10$ Hz). Our findings demonstrate that their performance increases with higher sensor resolution and approaches that of image ReID when observations are sufficiently dense. Our strongest network trained at the largest scale achieves ReID accuracy exceeding $90\%$ for rigid objects and $85\%$ for deformable objects (without any explicit skeleton normalization). To our knowledge, we are the first to study object re-identification from real point cloud observations.

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。

↑