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 2607.00529cs.CV
收藏

NoPA: 非参数化在线3D场景图生成

NoPA: Non-Parametric Online 3D Scene Graph Generation

Department of Computer Science, National University of Singapore(新加坡国立大学计算机科学系)

Qi Xun Yeo, Seungjun Lee, Yan Li, Gim Hee Lee

AI总结:

针对现有3D场景图生成方法无法实时运行的问题,提出NoPA,用非参数化分布表示物体,保留几何细节,并通过基于最大均值差异的合并策略和关系传播,实现实时且准确的在线场景图生成。

摘要:

经典的3D场景图生成方法由于环境映射的高计算成本和生成中间点云表示的需求,无法实时工作。为了缓解这一问题,最近的一项工作摒弃了点云,转而采用每个物体的轻量级高斯分布。这种近似大大加快了推理速度,实现了实时3D场景图生成。然而,该表示有两个关键弱点:1)每个物体被近似为单个3D高斯,导致3D几何细节严重丢失。2)这种近似与真实物体几何之间的差异加剧了在线推理过程中物体候选的不准确合并。为了解决这些问题,我们提出了NoPA,它将每个物体表示为单独的非参数化分布。这种表述保留了3D几何信息,同时保持了参数化高斯公式的实时推理能力。基于我们新颖的物体表示,我们提出了一种定制的合并策略来恢复连贯的物体实例。具体来说,我们利用核密度估计上的最大均值差异,在在线探索过程中实现物体候选的鲁棒合并,同时最小化增加的计算复杂度。关键在于为每个物体维护一个固定的粒子集。此外,为了纠正由错误分类物体引起的关系损失,NoPA在高亲和力物体之间传播关系。实验表明,NoPA在不牺牲实时推理速度的情况下,显著优于当前方法。

英文摘要:

Classic 3D scene graph generation approaches fail to work in real-time due to the heavy computational cost of environment mapping and the need to generate intermediate point-cloud representations. To alleviate this issue, a recent work eschews point clouds in favor of a lightweight Gaussian distribution for each object. This approximation drastically speeds up inference and enables real-time 3D scene graph generation. However, the representation has two key weaknesses. \textbf{1)} Each object is approximated by a single 3D Gaussian, which causes a severe loss of 3D geometric detail. \textbf{2)} The discrepancy between this approximation and the true object geometry exacerbates the inaccurate merging of object candidates during online inference. To address these issues, we propose \textbf{NoPA}, which represents each object as a separate non-parametric distribution. This formulation retains 3D geometric information while preserving real-time inference of the parametric Gaussian formulation. To build upon our novel object representation, we propose a tailored merging strategy to recover coherent object instances. Specifically, we leverage maximum mean discrepancy on kernel density estimates to enable robust merging of object candidates during online exploration while minimizing added computational complexity. The key is to maintain a fixed particle set per object. Furthermore, to rectify the relation loss caused by misclassified objects, NoPA propagates relationships between objects with high affinity. Experiments show that NoPA substantially outperforms current methods without sacrificing real-time inference speed.

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。

↑