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 2402.16174cs.CVcs.AIcs.RO
收藏

GenNBV:用于主动3D重建的泛化下一最佳视角策略

GenNBV: Generalizable Next-Best-View Policy for Active 3D Reconstruction

Shanghai AI Laboratory(上海人工智能实验室);The Chinese University of Hong Kong(香港中文大学)

Xiao Chen, Quanyi Li, Tai Wang, Tianfan Xue, Jiangmiao Pang

AI总结:

提出端到端泛化NBV策略GenNBV,采用强化学习扩展至5D动作空间并引入多源状态嵌入,在跨数据集未见物体上实现高覆盖率,显著提升主动3D重建性能。

摘要:

尽管神经辐射场领域的最新进展使得大规模场景的真实数字化成为可能,但图像捕捉过程仍然耗时且费力。先前的工作尝试使用下一最佳视角(NBV)策略来实现主动3D重建过程的自动化。然而,现有的NBV策略严重依赖手工设计的标准、受限的动作空间或逐场景优化的表示。这些限制条件限制了它们的跨数据集泛化能力。为了克服这些限制,我们提出了GenNBV,一种端到端的泛化NBV策略。我们的策略采用基于强化学习(RL)的框架,并将典型的受限动作空间扩展到5D自由空间。它使我们的智能体无人机能够从任意视角进行扫描,甚至在训练期间与未见过的几何形状进行交互。为了提升跨数据集泛化能力,我们还提出了一种新颖的多源状态嵌入,包括几何、语义和动作表示。我们使用Isaac Gym模拟器结合Houses3K和OmniObject3D数据集建立了一个基准来评估此NBV策略。实验表明,我们的策略在这些数据集的未见建筑级物体上分别实现了98.26%和97.12%的覆盖率,优于先前的解决方案。

英文摘要:

While recent advances in neural radiance field enable realistic digitization for large-scale scenes, the image-capturing process is still time-consuming and labor-intensive. Previous works attempt to automate this process using the Next-Best-View (NBV) policy for active 3D reconstruction. However, the existing NBV policies heavily rely on hand-crafted criteria, limited action space, or per-scene optimized representations. These constraints limit their cross-dataset generalizability. To overcome them, we propose GenNBV, an end-to-end generalizable NBV policy. Our policy adopts a reinforcement learning (RL)-based framework and extends typical limited action space to 5D free space. It empowers our agent drone to scan from any viewpoint, and even interact with unseen geometries during training. To boost the cross-dataset generalizability, we also propose a novel multi-source state embedding, including geometric, semantic, and action representations. We establish a benchmark using the Isaac Gym simulator with the Houses3K and OmniObject3D datasets to evaluate this NBV policy. Experiments demonstrate that our policy achieves a 98.26% and 97.12% coverage ratio on unseen building-scale objects from these datasets, respectively, outperforming prior solutions.

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。

↑