发表机构
Argonne National Laboratory; University of Illinois Chicago; Northwestern University（阿贡国家实验室; 芝加哥大学伊利诺伊分校; 西北大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对自主实验室机器人的神经三维重建，开展跨平台基准测试，评估NeRF、3D Gaussian Splatting及SAM3D的性能，为构建分层重建流水线提供依据。
AI 中文摘要
执行实验室任务的自主机器人依赖三维重建流水线，该流水线需在物理控制环路的延迟预算内，将原始相机数据流转化为可执行的物体表征。神经三维重建方法已展现出高质量的视图合成效果，但其在实验室机器人实际运行的计算平台上的实时可行性仍缺乏充分表征。本研究开展了神经三维重建方法的系统性计算平台基准测试，在从单板计算机到服务器级节点的各类带GPU的计算设备上，评估NeRF和3D Gaussian Splatting的训练与渲染性能，并将Meta公司的SAM3D单图像重建方法置于同一评估维度，量化其相对于逐场景优化方法的延迟与保真度差距。结果显示，Gaussian Splatting的渲染质量高于NeRF，但GPU成本更高；板载计算资源无法支撑全逐场景优化达到交互速率。对SAM3D的初步评估表明，其可在数秒内生成合理的物体几何结构，但存在细节不匹配问题，可能损害下游操作。综上，这些发现推动了分层流水线的发展，其中轻量级前馈重建可维持实验室机器人的实时感知与跟踪环路，而更繁重的神经重建则被选择性调度至合适的计算资源上。
英文摘要
Autonomous robots performing laboratory tasks depend on 3D reconstruction pipelines that can turn raw camera streams into actionable object representations within the latency budget of a physical control loop. Neural 3D reconstruction methods have demonstrated high-quality view synthesis, but their real-time viability across the compute platforms on which laboratory robots actually run remains poorly characterized. In this work, we present a systematic compute-platform benchmark of neural 3D reconstruction methods, evaluating NeRF and 3D Gaussian Splatting training and rendering on GPU-enabled computing devices ranging from single-board computers to server-class nodes, and place Meta's SAM3D single-image reconstruction on the same axes to quantify its latency and fidelity gap relative to per-scene optimization. Our results show that Gaussian Splatting yields higher rendering quality than NeRF at greater GPU cost, and that onboard compute is insufficient for full per-scene optimization at interactive rates. Our preliminary assessment on SAM3D indicates that it delivers plausible object geometry within seconds, but with detail mismatches that can compromise downstream manipulation. Together, these findings motivate tiered pipelines in which lightweight feed-forward reconstruction sustains the real-time perception-and-tracking loop for laboratory robots, while heavier neural reconstruction is scheduled selectively on suitable compute.
CommentsThis manuscript is peer-reviewed from the committees in the workshop "VAxAutoSci: Visual Analytics in the Age of Autonomous Scientific Discovery" in conjunction with 2026 IEEE Visualization & Visual Analytics
