[论文] Post-Training Semantic Lifting for 3D Gaussian Splatting: Separating D...

论文概要

研究领域: CV
作者: Iván Verdugo Guerra, Ezequiel López Rubio, Jorge García González
发布时间: 2026-10-06
arXiv: 2610.08756

中文摘要

3D高斯溅射模型的同一个高斯从许多视角观察，这些视角并不总是在它所属的类别上达成一致。高斯可能在某些视图中被遮挡，检测器的置信度从一个视图到另一个视图也不相同。另一方面，真值以标注网格的形式给出，因为两次训练运行不会产生相同的高斯。本工作提出了一种后训练提升方法，一次处理一个目标类别，并结合来自所有视图的信息。目标和非目标证据同时累积，按每个高斯在每个视图中的可见性加权。之后，用两个阈值过滤高斯：主阈值β选择高置信度种子，较低的阈值γβ在其周围添加连通分量。评估时，标签从高斯转移到既可见又被标注的网格顶点。通过这种设计，我们可以分离三种错误来源：2D检测器、提升和表示之间的转移。验证场景上的平均mIoU在使用数据集标注的掩码时为0.93，使用YOLO掩码时为0.65；在ScanNet++测试场景上分别为0.80和0.54。与之前版本对每个视图的证据进行阈值处理相比，该改进将测试mIoU提高了0.24，并使得对两个数据集的所有类别和场景使用单一阈值成为可能。错误分析表明大部分剩余错误来自检测器。

原文摘要

The same Gaussian of a 3D Gaussian Splatting model is seen from many views, and these views do not always agree on the class it belongs to. The Gaussian may be occluded in some of them, and the confidence of the detector is not the same from one view to another. The ground truth, on the other hand, is given as an annotated mesh, because two training runs do not produce the same Gaussians. In this work, we propose a post-training lifting method that works with one target class at a time and combines the information coming from all the views. Target and non-target evidence are accumulated simultaneously, weighted by the visibility of each Gaussian in each view. After that, the Gaussians are filtered with two thresholds: a main threshold β selects the high-confidence seeds, and a lower one γβ...

自动采集于 2026-10-08

#论文 #arXiv #CV #小凯