MSG-Loc: Multi-Label Likelihood-based Semantic Graph Matching for Object-Level Global Localization

2025-12-03

AI Summary

In global robot localization under unknown object categories and semantically ambiguous conditions, erroneous classification and incorrect data association lead to pose estimation bias. To address this, we propose a multi-label likelihood-driven semantic graph matching framework. Our method constructs multi-label semantic graphs to explicitly model object observations and their contextual relationships, and introduces a context-aware likelihood propagation mechanism that enhances semantic matching robustness via maximum-likelihood fusion between nodes and their neighborhoods. The framework supports both closed-set and open-set detection configurations, ensuring scalability to large-vocabulary object categories. Experiments on real indoor and synthetic datasets demonstrate that our approach significantly reduces mismatch rates under high-ambiguity conditions, thereby improving global localization accuracy and robustness. It is particularly effective for large-scale semantic localization in open, complex environments.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Robots are often required to localize in environments with unknown object classes and semantic ambiguity. However, when performing global localization using semantic objects, high semantic ambiguity intensifies object misclassification and increases the likelihood of incorrect associations, which in turn can cause significant errors in the estimated pose. Thus, in this letter, we propose a multi-label likelihood-based semantic graph matching framework for object-level global localization. The key idea is to exploit multi-label graph representations, rather than single-label alternatives, to capture and leverage the inherent semantic context of object observations. Based on these representations, our approach enhances semantic correspondence across graphs by combining the likelihood of each node with the maximum likelihood of its neighbors via context-aware likelihood propagation. For rigorous validation, data association and pose estimation performance are evaluated under both closed-set and open-set detection configurations. In addition, we demonstrate the scalability of our approach to large-vocabulary object categories in both real-world indoor scenes and synthetic environments.

Problem

• Addresses object misclassification in semantic global localization
• Reduces incorrect associations via multi-label graph matching
• Enhances semantic correspondence with context-aware likelihood propagation

Innovation

• Multi-label graph representations capture semantic context
• Context-aware likelihood propagation enhances semantic correspondence
• Scalable to large-vocabulary object categories in diverse environments