Feature Suggestion: Semantic Search for Intelligent Photo Retrieval

[snehitdesu]

Hi, I’ve been exploring PictoPy and really liked its approach to privacy-focused AI photo management.

I would like to suggest adding a semantic search feature to improve image retrieval.

Currently, search is mainly based on tags and face recognition. Adding semantic search would allow users to search images using natural language queries like:

• "beach sunset"
• "friends at party"
• "dog playing in park"

Proposed approach:

• Use an embedding model like CLIP to generate vector representations of images
• Store embeddings locally in the database
• Perform similarity search for user queries
• Integrate with existing tagging and face recognition system

This feature can significantly enhance user experience while keeping everything local and privacy-friendly.

I would love to explore contributing to this feature.

Thanks!

[github-actions]

👋 Thanks for opening this issue! The maintainers will review it shortly. Till then, do not open any PR. This is a strict rule we like to follow around here :)

[snehitdesu]

Hi, I have also included this idea in my GSoC proposal and explored a basic prototype for image processing.

I would like to start working on this feature step by step. Please let me know if this direction is correct or if I should modify the approach.

Thanks!

[JITENDRAGAHERWAR03]

Hello 👋
I am interested in working on this issue. I have basic knowledge of Python and I’m eager to learn and contribute. I will try to provide a clean and well-documented solution.
Please assign this issue to me if possible.
Thank you 😊