This personal learning project seeks to merge photography, software engineering, and family memories by creating an interactive experience to explore a large collection of family photos. The system will feature smart search capabilities focused on privacy, running entirely on local hardware. Key goals include automatic identification and tagging of family members through local face recognition, generating descriptive captions, and enabling natural language searches like “Show me photos of us at the beach in Luquillo from last summer.” A high-level project plan has been developed with the help of AI tools, and the creator seeks insights from others who have built similar projects. They are particularly interested in advice regarding the integration of structured metadata (EXIF), face recognition data, and semantic vector search into a cohesive application. Recommendations on tools, models, and best practices are welcome for this 2025 project.
Source link
Ask HN: What’s the Ideal Tech Stack for a Self-Hosted Photo Library with Local AI in 2025?
Share
Read more