BriefGPT.xyz
Sep, 2023
SGRec3D: 自主学习的3D场景图通过对象级场景重建
SGRec3D: Self-Supervised 3D Scene Graph Learning via Object-Level Scene Reconstruction
HTML
PDF
Sebastian Koch, Pedro Hermosilla, Narunas Vaskevicius, Mirco Colosi, Timo Ropinski
TL;DR
提出了SGRec3D,这是一种用于预先训练三维场景图预测的新方法,无需对象关系标签,可以利用大规模三维场景理解数据集进行训练。在细调期间只使用10%标记数据集便可优于未进行预先训练的模型。
Abstract
In the field of
3d scene understanding
,
3d scene graphs
have emerged as a new scene representation that combines geometric and semantic information about objects and their relationships. However, learning semanti
→