arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2404.05136cs.CVcs.AI
收藏

基于路径一致性的自监督多目标跟踪

Self-Supervised Multi-Object Tracking with Path Consistency

AWS AI Labs(AWS AI实验室);Northeastern University(东北大学)

Zijia Lu, Bing Shuai, Yanbei Chen, Zhenlin Xu, Davide Modolo

AI总结:

本文提出路径一致性自监督学习方法,通过让模型在不同跳帧观察路径下产生一致关联结果来训练目标匹配,无需人工身份标注,在多个跟踪数据集上超越现有无监督方法并接近有监督性能。

摘要:

本文提出了一种新颖的路径一致性概念,用于在不使用人工目标身份监督的情况下学习鲁棒的目标匹配。我们的核心思想是:为了跨帧跟踪一个目标,可以通过改变模型可观察的帧(即在观察中跳过某些帧)来从模型中获得多个不同的关联结果。由于观察的差异不会改变目标的身份,所获得的关联结果应当是一致的。基于这一原理,我们生成多条观察路径,每条路径指定一组不同的被跳过帧,并构建路径一致性损失(Path Consistency Loss),以强制不同观察路径下的关联结果保持一致。我们使用所提出的损失仅以自监督方式训练目标匹配模型。通过在三个跟踪数据集(MOT17、PersonPath22、KITTI)上的大量实验,我们证明该方法在各种评估指标上以一致的幅度优于现有的无监督方法,甚至达到了接近有监督方法的性能。

英文摘要:

In this paper, we propose a novel concept of path consistency to learn robust object matching without using manual object identity supervision. Our key idea is that, to track a object through frames, we can obtain multiple different association results from a model by varying the frames it can observe, i.e., skipping frames in observation. As the differences in observations do not alter the identities of objects, the obtained association results should be consistent. Based on this rationale, we generate multiple observation paths, each specifying a different set of frames to be skipped, and formulate the Path Consistency Loss that enforces the association results are consistent across different observation paths. We use the proposed loss to train our object matching model with only self-supervision. By extensive experiments on three tracking datasets (MOT17, PersonPath22, KITTI), we demonstrate that our method outperforms existing unsupervised methods with consistent margins on various evaluation metrics, and even achieves performance close to supervised methods.

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。

↑