pith. sign in

arxiv: 2008.09506 · v1 · pith:BYL3KUXBnew · submitted 2020-08-20 · 💻 cs.CV · cs.LG· cs.MA· cs.MM· cs.RO

Graph Neural Networks for 3D Multi-Object Tracking

classification 💻 cs.CV cs.LGcs.MAcs.MMcs.RO
keywords featuresfeatureproposeaffinityassociationdatagraphindependently
0
0 comments X
read the original abstract

3D Multi-object tracking (MOT) is crucial to autonomous systems. Recent work often uses a tracking-by-detection pipeline, where the feature of each object is extracted independently to compute an affinity matrix. Then, the affinity matrix is passed to the Hungarian algorithm for data association. A key process of this pipeline is to learn discriminative features for different objects in order to reduce confusion during data association. To that end, we propose two innovative techniques: (1) instead of obtaining the features for each object independently, we propose a novel feature interaction mechanism by introducing Graph Neural Networks; (2) instead of obtaining the features from either 2D or 3D space as in prior work, we propose a novel joint feature extractor to learn appearance and motion features from 2D and 3D space. Through experiments on the KITTI dataset, our proposed method achieves state-of-the-art 3D MOT performance. Our project website is at http://www.xinshuoweng.com/projects/GNN3DMOT.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.