English

VL-KnG: Persistent Spatiotemporal Knowledge Graphs from Egocentric Video for Embodied Scene Understanding

Robotics 2026-03-25 v2 Artificial Intelligence

Abstract

Vision-language models (VLMs) demonstrate strong image-level scene understanding but often lack persistent memory, explicit spatial representations, and computational efficiency when reasoning over long video sequences. We present VL-KnG, a training-free framework that constructs spatiotemporal knowledge graphs from monocular video, bridging fine-grained scene graphs and global topological graphs without 3D reconstruction. VL-KnG processes video in chunks, maintains persistent object identity via LLM-based Spatiotemporal Object Association (STOA), and answers queries via Graph-Enhanced Retrieval (GER), a hybrid of GraphRAG subgraph retrieval and SigLIP2 visual grounding. Once built, the knowledge graph eliminates the need to re-process video at query time, enabling constant-time inference regardless of video length. Evaluation across three benchmarks, OpenEQA, NaVQA, and WalkieKnowledge (our newly introduced benchmark), shows that VL-KnG matches or surpasses frontier VLMs on embodied scene understanding tasks at significantly lower query latency, with explainable, graph-grounded reasoning. Real-world robot deployment confirms practical applicability with constant-time scaling.

Keywords

Cite

@article{arxiv.2510.01483,
  title  = {VL-KnG: Persistent Spatiotemporal Knowledge Graphs from Egocentric Video for Embodied Scene Understanding},
  author = {Mohamad Al Mdfaa and Svetlana Lukina and Timur Akhtyamov and Arthur Nigmatzyanov and Dmitrii Nalberskii and Sergey Zagoruyko and Gonzalo Ferrer},
  journal= {arXiv preprint arXiv:2510.01483},
  year   = {2026}
}
R2 v1 2026-07-01T06:11:59.149Z