PhotoGraph: Claim-Centric Knowledge Graphs for Personal Photo Management

Omair Shahzad Bhatti, Abdulrahman Mohamed Selim, László Kopácsi, Michael Barz, Daniel Sonntag

Track: Position & Demo

Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.

Student Paper: No

Generative AI Compliance And Declaration: I confirm that this submission complies with GenAIK-NORA Generative AI Policy. The manuscript includes a mandatory declaration section explicitly stating whether Generative AI was used and, if applicable, describing the extent of its usage.

Keywords: Personal photo management, Knowledge graphs, Claim-centric representation, Contextual image retrieval, Human-AI collaboration, Visual storytelling, Photobook co-creation

TL;DR: PhotoGraph turns personal photos into an editable, evidence-grounded knowledge graph, enabling transparent contextual search and fact-based photobook storytelling.

Abstract: Personal photo collections are increasingly organised with machine learning, yet many systems still cannot answer contextual questions in an inspectable way. Embedding-based retrieval supports natural language search, but provides weak support for relational constraints, visual evidence, provenance, and persistent correction. We present \tool{}, a claim-centric, spatially and temporally aware knowledge graph framework for personal photo understanding and photobook co-creation. \tool{} represents model outputs as evidence-grounded claims with lifecycle state, enabling contextual retrieval with justifications and event-based creation with fact-grounded storylines.

Submission Number: 6