{"type":"ScholarlyArticle","title":"Riemannian Graph Convolutional Network for Skeleton-Based Two-Person Interaction Recognition","authors":[{"name":"Rui Wang","slug":"rui-wang","given":"Rui","family":"Wang","affiliation":""},{"name":"Zihao Bi","slug":"zihao-bi","given":"Zihao","family":"Bi","affiliation":""},{"name":"Chen Hu","slug":"chen-hu","given":"Chen","family":"Hu","affiliation":""},{"name":"Xiaoning Song","slug":"xiaoning-song","given":"Xiaoning","family":"Song","affiliation":""},{"name":"Xiao-Jun Wu","slug":"xiao-jun-wu","given":"Xiao-Jun","family":"Wu","affiliation":""},{"name":"Nicu Sebe","slug":"nicu-sebe","given":"Nicu","family":"Sebe","affiliation":""},{"name":"Ziheng Chen","slug":"ziheng-chen","given":"Ziheng","family":"Chen","affiliation":""}],"abstract":"In the field of skeleton-based human action recognition, Graph Convolutional Networks (GCNs) have become a dominant framework. However, existing GCN-based approaches often treat the sequences of two-person interaction as separate entities, ignoring the inherent semantic dependencies and spatial correlations between interacting subjects. Furthermore, high-order skeleton representations naturally exhibit non-Euclidean structures, where Euclidean deep learning models are inherently limited in explicitly capturing and preserving such geometric information. As a countermeasure, we propose a Riemannian Graph Convolutional Network (RGCN) that operates on the Symmetric Positive Definite (SPD) manifolds. Specifically, we model high-order skeletal statistics via Gaussian embedding and propose a Riemannian network to capture inter-subject interactions and global correlations. The proposed RGCN is instantiated under three SPD geometries, and its effectiveness is validated through extensive experiments on three interaction benchmarks. Extensive experimental results show that RGCN provides a competitive and geometrically grounded alternative for skeleton-based interaction recognition.","doi":"10.24963/ijcai.2026/196","url":"https://ijcai.org/proceedings/2026/196","pdf":"https://ijcai.org/proceedings/2026/0196.pdf","markdown":"https://ijcai.org/proceedings/2026/196.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/196","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Computer Vision","track":"Main Track","number":196,"pages":{"first":1758,"last":1766},"keywords":["Machine Learning: Geometric learning","Machine Learning: Deep learning architectures","Computer Vision: Action and behavior recognition","Computer Vision: Machine learning for vision"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}