Weakly-Supervised 3D Scene Graph Generation via Visual-Linguistic Assisted Pseudo-labeling

arXiv:2404.02527v1 Announce Type: new
Abstract: Learning to build 3D scene graphs is essential for real-world perception in a structured and rich fashion. However, previous 3D scene graph generation methods utilize a fully supervised learning manner and require a large amount of entity-level annotation data of objects and relations, which is extremely resource-consuming and tedious to obtain. To tackle this problem, we propose 3D-VLAP, a weakly-supervised 3D scene graph generation method via Visual-Linguistic Assisted Pseudo-labeling. Specifically, our 3D-VLAP exploits the superior …

abstract annotation arxiv build cs.cv data fashion graph graphs however labeling objects perception relations supervised learning type via visual weakly-supervised world