Sparse3DPR: Training-Free 3D Hierarchical Scene Parsing and Task-Adaptive Subgraph Reasoning from Sparse RGB Views

TL;DR

Sparse3DPR is a training-free 3D scene understanding framework that uses hierarchical scene graphs and subgraph extraction to improve reasoning accuracy and efficiency from sparse RGB views, leveraging pre-trained LLMs.

Contribution

It introduces a hierarchical plane-enhanced scene graph and task-adaptive subgraph extraction, enabling open-vocabulary reasoning and dynamic noise filtering without training.

Findings

28.7% EM@1 improvement over ConceptGraphs

78.2% speedup compared to ConceptGraphs

Achieves comparable performance to training-based methods on ScanQA

Abstract

Recently, large language models (LLMs) have been explored widely for 3D scene understanding. Among them, training-free approaches are gaining attention for their flexibility and generalization over training-based methods. However, they typically struggle with accuracy and efficiency in practical deployment. To address the problems, we propose Sparse3DPR, a novel training-free framework for open-ended scene understanding, which leverages the reasoning capabilities of pre-trained LLMs and requires only sparse-view RGB inputs. Specifically, we introduce a hierarchical plane-enhanced scene graph that supports open vocabulary and adopts dominant planar structures as spatial anchors, which enables clearer reasoning chains and more reliable high-level inferences. Furthermore, we design a task-adaptive subgraph extraction method to filter query-irrelevant information dynamically, reducing contextual noise and improving 3D scene reasoning efficiency and accuracy. Experimental results demonstrate the superiority of Sparse3DPR, which achieves a 28.7% EM@1 improvement and a 78.2% speedup compared with ConceptGraphs on the Space3D-Bench. Moreover, Sparse3DPR obtains comparable performance to training-based methods on ScanQA, with additional real-world experiments confirming its robustness and generalization capability.