提升实时连续3D重建速度，效率提高1680倍且保持精度。
ImprovedVBGS: Real-time Continual Variational Bayes Gaussian Splatting
- 通过空间截断推断与优化重分配加速推断过程。
- 在RTX 3070 Ti上每帧延迟从84秒降至0.05秒，提升1680倍。
- 无需额外训练成本，即可改善新视角生成质量。
即时重建是机器人与自主导航等领域的重要需求。变分贝叶斯高斯点阵（VBGS）利用坐标上升变分推断（CAVI）实现无回放缓冲的持续学习，但其对所有观测点进行逐帧迭代导致速度过慢，难以满足严格内存与延迟要求。本文提出ImprovedVBGS，一种加速的即时持续重建框架，主要通过(1)空间截断变分推断，(2)改进的重分配机制（包含前向传播、截断及消除无效动态重新编译）。在NeRF合成数据集上，我们将在RTX 3070 Ti上的平均每帧延迟从约84.0秒降至约0.050秒，实现1680倍加速，同时保持重建质量。此外，使用精确渲染器进一步提升了新视角合成质量，且无需增加训练成本。
原文摘要 · Abstract (English)
On-the-fly reconstruction is a key requirement for many applications in robotics and autonomous navigation. Variational Bayes Gaussian Splatting (VBGS) enables continual learning without replay buffers using Coordinate Ascent Variational Inference (CAVI), but its per-frame iterations over all observed points make it too slow for real-time use with strict memory and latency requirements. We present ImprovedVBGS, an accelerated framework for on-the-fly continual reconstruction. This is achieved primarily through (i) spatially truncated variational inference, and (ii) improved reassignment that uses forwarding, truncation and eliminates wasteful dynamic recompilation. On the NeRF synthetic dataset, we reduce mean per-frame latency from ~84.0 s to ~0.050 s on an RTX 3070 Ti, a 1680x speed-up while maintaining reconstruction quality. We also improve novel-view synthesis quality using an exact renderer with no added training costs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。