DeWorldSG: Depth-Aware 3D Semantic Scene Graph Generation via World-Model Priors

MCML Authors

Abdelrahman Elskhawy

→ Group Benjamin Busam
 Photogrammetry and Remote Sensing

Benjamin Busam

Prof. Dr.

Core PI

Photogrammetry and Remote Sensing

Abstract

resent DeWorldSG, a novel framework that generates spatio-temporally robust 3D Semantic Scene Graphs from RGB-D sequences. Existing methods often struggle to construct reliable 3D scene graphs due to unstable 3D object representations and missing relations caused by frame-wise inference. DeWorldSG addresses these issues by estimating instance-level geometric 3D Gaussian distributions through depth-guided filtering and representing each object as a probabilistic 3D node rather than a single projected point. To mitigate relational sparsity from frame-wise inference, our framework further aggregates spatiotemporal evidence across object pairs and refines relations using contextual priors derived from a world model (V-JEPA 2). Experiments on the 3DSSG and ReplicaSSG datasets demonstrate state-of-the-art (SoTA) performance in both object and predicate prediction, while producing temporally consistent scene structures. In particular, our method improves triplet recall by 77.4% and predicate recall by 23.2% over prior SoTA approaches, making it suitable for robotic manipulation and AR applications. Our code and models are open-sourced.

inproceedings KEH+26

---

ECCV 2026

19th European Conference on Computer Vision. Malmö, Sweden, Sep 08-12, 2026. To be published. Preprint available.

---

Authors

S.-Y. Kim • A. Elskhawy • T. Ha • D. Kim • E. Shin • B. Busam • W. Woo

---

Links

arXiv GitHub

---

Research Area

 B1 | Computer Vision