Photometoria is an AI-powered metadata generation system for photography that integrates local AI models with image organization software, currently Adobe Lightroom Classic. The system is designed for manual batch processing of large photo collections, using a multi-level tagging approach that analyzes individual photos for fine details, groups photos for contextual macro categories, incorporates user-provided context hints, and extracts EXIF metadata.
The project aims to automate photo tagging workflows for photographers managing substantial image libraries.
The project consists of:
- REST API : Backend service that orchestrates image analysis using local AI models (Ollama)
- Lightroom Plugin : Lua extension for Adobe Lightroom Classic that allows photographers to send images to the API and receive generated metadata directly in their catalog
- Testing Scripts : Python tools for testing and validating AI models
The main goal is to automate the photographic keywording process, drastically reducing the time required to catalog images while maintaining high quality and precision in the generated metadata.
Current Version: 0.3.0
Catalog-scoped architecture and multi-language AI analysis:
- Catalog identity — Projects and activities are now scoped to the Lightroom catalog; the plugin generates and persists a unique catalog UUID; all API endpoints are prefixed with
/api/catalogs/{catalog_id} - Multi-language support — Language selection in the New Activity dialog; AI tag generation in English, Italian, Spanish, Arabic, Chinese, and more
- Structured JSON tags — AI models return structured JSON for reliable tag parsing and downstream processing
- Storage resilience — Filesystem consistency checks on startup, quarantine for corrupt data, atomic file writes, partial recovery for malformed entries
- Input validations — Character limits for project context, photo upload constraints (size, MIME type, storage quota), model availability checks, disallow activities without photos
- Photo management — Upsert semantics on re-upload with replace confirmation, photo subset selection for activity creation, Remove Photos command in plugin
- UX improvements — Recent hosts list, real-time ETA for running activities, pre-select last used provider/model, configurable server name display, option to remove existing keywords on apply
Full-featured Lightroom Classic plugin with complete task and job management:
- Task management — Create, delete, edit context, rename tasks directly from Lightroom
- Photo workflow — Select, validate, export and upload photos to tasks; view task photos in Library
- Job lifecycle — Start AI analysis jobs, monitor real-time progress, cancel, retry, delete jobs
- Keyword application — Apply AI-generated tags as Lightroom keywords with confirmation dialog
- API enhancements — Server info endpoint, provider/model listing, client-side photo reconciliation, task-job relationship exposure
- REST API server with Axum framework
- Ollama integration for local AI inference
- Worker pool for GPU-bound job processing
- Job endpoints (CRUD, retry, cancel)
- Python testing scripts for model validation
The project is in active development.
Hardware:
-
NVIDIA GPU(s) recommended for AI model inference
-
Sufficient storage for photo uploads (configurable quota)
Software:
-
Ollama - Local AI model inference engine
-
Rust toolchain (for API development)
-
Python 3.11+ (for testing scripts)
- Install and start Ollama:
# Install Ollama (see https://ollama.ai)
ollama serve
# Pull recommended model
ollama pull qwen3-vl:8b- Test with Python scripts:
cd scripts
pip install -r requirements.txt
python3 test_models.py- Build and run API:
cd api
cargo build --release
cargo run --release- API Documentation - Comprehensive REST API specification and implementation guide
- Plugin Documentation - Lightroom Classic plugin integration
- Scripts Documentation - Python testing tools and usage
- Contributing Guidelines - Coding standards and guidelines for development
This project includes Claude Code skills for AI-assisted development. Skills provide on-demand development guidelines scoped to each component:
| Skill | Command | Description |
|---|---|---|
| API Development | /api-dev |
Rust code style, testing guidelines, tech stack, dev commands |
| Plugin Development | /plugin-dev |
LrView UI constraints, Lua testing, Lightroom workflow |
Invoke a skill at the start of a session when working on the corresponding component.
This project uses AI-assisted development with coding agents. Before contributing:
- Read the Contributing Guidelines for code style and conventions
- Review the API Documentation for architecture details
- Run tests and formatting before committing:
# Rust
cd api
cargo fmt && cargo clippy && cargo test
# Python
cd scripts
# Run your testsphotometoria/
├── api/ # Rust REST API server (Axum)
├── plugin/ # Lightroom Classic plugin (Lua)
├── scripts/ # Python testing and validation tools
├── CONTRIBUTING.md # Development and coding guidelines
└── README.md # This file
- Use imperative mood in commit messages
- Keep commits atomic (one logical change per commit)
- Run quality checks before committing
Copyright (c) 2026 Fabrizio Di Giuseppe
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
Trademark Notice: "Photometoria" is a trademark of Fabrizio Di Giuseppe.