A method transfers trajectories across 3D scenes by clustering objects, predicting hierarchical smooth maps from foundation model features, assembling them combinatorially, and refining for coherence.
SG-PGM: Partial graph matching network with semantic geometric fusion for 3D scene graph alignment and its downstream tasks
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
A learned end-to-end differentiable method for hierarchical scene graph matching outperforms combinatorial baselines in F1 score and speed for BIM-assisted robot localization, with zero-shot generalization from floor-plan training to real LiDAR data.
citing papers explorer
-
Analogical Trajectory Transfer
A method transfers trajectories across 3D scenes by clustering objects, predicting hierarchical smooth maps from foundation model features, assembling them combinatorially, and refining for coherence.
-
Learning-Based Hierarchical Scene Graph Matching for Robot Localization Leveraging Prior Maps
A learned end-to-end differentiable method for hierarchical scene graph matching outperforms combinatorial baselines in F1 score and speed for BIM-assisted robot localization, with zero-shot generalization from floor-plan training to real LiDAR data.