外观
论文arXiv 2610.2325012 天前
LonGaussian：用于长上下文前馈 Gaussian 重建的基础模型先验跟踪
原文链接：LonGaussian: Tracking Foundation Priors for Long-Context Feedforward Gaussian Reconstruction（arXiv 2610.23250 · 发布日期：2026-10-16）
LonGaussian: Tracking Foundation Priors for Long-Context Feedforward Gaussian Reconstruction
LonGaussian面向长图像序列重建，结合流式前馈框架、点轨迹合并与深度一致性过滤，在DL3DV和RealEstate10K验证。
核心要点
- 长视频3D场景重建对机器人感知重要，但全视角联合处理耗显存，像素对齐预测易冗余。
- 方法基于预训练流式3D重建骨干，将长上下文几何特征解码为紧凑、全局可渲染的Gaussian场景。
- 在每个分块内用多帧点轨迹关联同一表面观测，并通过学习式合并器整合对应Gaussian预测。
- 在DL3DV和RealEstate10K上，相比评估的前馈基线提升新视角渲染质量，并使用更少Gaussian。
验证范围与局限：验证范围为DL3DV和RealEstate10K基准上的新视角渲染质量与Gaussian数量。
英文摘要 (Abstract)
Reconstructing high-quality, renderable 3D scenes from long videos is essential for robotic perception and embodied intelligence. Many feed-forward Gaussian reconstruction methods rapidly predict 3D scenes from a small set of images, but extending them to longer sequences presents two bottlenecks: jointly processing all views increases GPU memory requirements, while accumulating pixel-aligned predictions redundantly represents the same surfaces and increases storage and rendering costs. We present LonGaussian, a streaming feed-forward framework that reconstructs compact, globally renderable Gaussian scenes from long image sequences with known intrinsics and unknown camera poses. Built on a pretrained streaming 3D reconstruction backbone, LonGaussian decodes long-context geometric features into dense Gaussians. Within each chunk, multi-frame point tracks link repeated observations of the same surface and guide a learned merger to consolidate the corresponding Gaussian predictions. Rendered-depth consistency filtering further suppresses Gaussians from other chunks that incorrectly occlude locally reconstructed surfaces. On DL3DV and RealEstate10K benchmarks, LonGaussian improves novel-view rendering quality while using substantially fewer Gaussians than the evaluated feed-forward baselines.
作者：Zhoufeng Xie, Guanjun Wu, Shuai Zhang, Xinggang Wang, Wenyu Liu
来源
本页事实对应的公开页面，以原文为准。