Presentation

<
LLM-enhanced Scene Graph Learning for Household Rearrangement
SessionLayouts & Procedural
DescriptionWe present an LLM-enhanced scene graph framework for autonomous household rearrangement. By transforming scene graphs into affordance-enhanced graphs that encode scene-grounded object functionality and user preferences, our method detects misplaced objects and plans correct placements. We demonstrate state-of-the-art performance with an end-to-end robotic system in simulated and real-world environments.
Authors
Wenhao Li
National University of Defense Technology
Shilong Zou
National University of Defense Technology
Zhinan Yu
National University of Defense Technology
Zheng Zhou
National University of Defense Technology
Wenxuan Li
National University of Defense Technology
Chenyang Zhu
National University of Defense Technology
Ruizhen Hu
Shenzhen University
Kai Xu
Institute of AI for Industries Chinese Academy of Science
Presenter
Wenhao Li
National University of Defense Technology (NUDT)
Event Type
Technical Paper
TimeTuesday, 21 July 20269:00am - 9:10am PDT
LocationRoom 403 A
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Tuesday, 21 July 20269:00am - 10:30am PDTRoom 403 A
Keywords
Geometry
Modeling
Registration Categories
Full Conference Supporter
Full Conference
Next PresentationNext PresentationCasLayout: Cascaded 3D Layout Diffusion for Indoor Scene Synthesis with Implicit Relation Modeling
Similar Presentations
CasLayout: Cascaded 3D Layout Diffusion for Indoor Scene Synthesis with Implicit Relation Modeling
NVIDIA Hands-On Lab: Isaac Sim and the Fine Art of Making Your Robot Pick Things Up