arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.02129cs.CV
收藏

GSOT3D:迈向真实场景中的通用三维单目标跟踪

GSOT3D: Towards Generic 3D Single Object Tracking in the Wild

University of Chinese Academy of Sciences(中国科学院大学);Institute of Software Chinese Academy of Sciences(中国科学院软件研究所);University of North Texas(北得克萨斯大学)

Yifan Jiao, Yunhao Li, Junhua Ding, Qing Yang, Song Fu, Heng Fan, Libo Zhang

AI总结:

该研究构建了多模态、大规模真实场景三维单目标跟踪基准 GSOT3D,评估八类点云跟踪器,并提出性能领先的渐进式时空跟踪器 PROT3D。

摘要:

本文提出了一个名为 GSOT3D 的新型基准,旨在促进真实场景中通用三维单目标跟踪(SOT)的发展。具体而言,GSOT3D 提供620个序列、123K帧,并覆盖54个对象类别。每个序列都提供多种模态,包括点云(PC)、RGB图像和深度图。这使 GSOT3D 能够支持多种三维跟踪任务,例如基于点云的单模态三维 SOT,以及基于 RGB-PC 或 RGB-D 的多模态三维 SOT,从而极大拓宽了三维目标跟踪的研究方向。为提供高质量的逐帧三维标注,所有序列均经过人工标注,并接受多轮细致检查与精修。据我们所知,GSOT3D 是致力于各类通用三维目标跟踪任务的最大基准。为了解现有三维跟踪器的表现,并为未来在 GSOT3D 上的研究提供比较,我们评估了八个具有代表性的基于点云的跟踪模型。评估结果表明,这些模型在 GSOT3D 上性能显著下降,要实现鲁棒且通用的三维目标跟踪仍需更多努力。此外,为鼓励未来研究,我们提出了一个简单而有效的通用三维跟踪器 PROT3D,它通过渐进式时空网络定位目标对象,并以较大幅度优于所有当前方案。通过发布 GSOT3D,我们期望推动未来三维跟踪研究与应用的进一步发展。我们的基准、模型以及评估结果将在网页 https://github.com/ailovejinx/GSOT3D 公开发布。

英文摘要:

In this paper, we present a novel benchmark, GSOT3D, that aims at facilitating development of generic 3D single object tracking (SOT) in the wild. Specifically, GSOT3D offers 620 sequences with 123K frames, and covers a wide selection of 54 object categories. Each sequence is offered with multiple modalities, including the point cloud (PC), RGB image, and depth. This allows GSOT3D to support various 3D tracking tasks, such as single-modal 3D SOT on PC and multi-modal 3D SOT on RGB-PC or RGB-D, and thus greatly broadens research directions for 3D object tracking. To provide highquality per-frame 3D annotations, all sequences are labeled manually with multiple rounds of meticulous inspection and refinement. To our best knowledge, GSOT3D is the largest benchmark dedicated to various generic 3D object tracking tasks. To understand how existing 3D trackers perform and to provide comparisons for future research on GSOT3D, we assess eight representative point cloud-based tracking models. Our evaluation results exhibit that these models heavily degrade on GSOT3D, and more efforts are required for robust and generic 3D object tracking. Besides, to encourage future research, we present a simple yet effective generic 3D tracker, named PROT3D, that localizes the target object via a progressive spatial-temporal network and outperforms all current solutions by a large margin. By releasing GSOT3D, we expect to advance further 3D tracking in future research and applications. Our benchmark and model as well as the evaluation results will be publicly released at our webpage https://github.com/ailovejinx/GSOT3D.

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。

↑