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 2301.08147cs.CVcs.RO
收藏

基于RGB-D的分类对象姿态和形状估计:方法,数据集和评估

RGB-D-Based Categorical Object Pose and Shape Estimation: Methods, Datasets, and Evaluation

KTH Royal Institute of Technology(KTH皇家理工学院)

Leonard Bruns, Patric Jensfelt

摘要:

最近,已经提出了用于在每个类别级别上对对象进行6D姿态和形状估计的各种方法。这项工作提供了该领域的方法,数据集和评估协议的概述。首先,概述了现有的工作和他们的共性和差异。其次,我们对主要的评估协议进行了批判性的研究,包括指标和数据集。基于这些发现,我们提出了一组新的指标,为Redwood数据集提供了新的注释,并在公平的比较中评估了最先进的方法。结果表明,现有的方法不能很好地推广到无约束的方向,实际上是严重偏向于直立的对象。我们提供了一个易于使用的评估工具箱,其中包含定义良好的指标、方法和数据集接口,可以与各种最先进的方法进行评估和比较(https://github.com/roym899/pose_and_shape_evaluation)。

英文摘要:

Recently, various methods for 6D pose and shape estimation of objects at a per-category level have been proposed. This work provides an overview of the field in terms of methods, datasets, and evaluation protocols. First, an overview of existing works and their commonalities and differences is provided. Second, we take a critical look at the predominant evaluation protocol, including metrics and datasets. Based on the findings, we propose a new set of metrics, contribute new annotations for the Redwood dataset, and evaluate state-of-the-art methods in a fair comparison. The results indicate that existing methods do not generalize well to unconstrained orientations and are actually heavily biased towards objects being upright. We provide an easy-to-use evaluation toolbox with well-defined metrics, methods, and dataset interfaces, which allows evaluation and comparison with various state-of-the-art approaches (https://github.com/roym899/pose_and_shape_evaluation).

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。

↑