Package: ai-foto

All packages Open in Cursor /Users/friedrichstr/ai-mac-studio/agents/media/ai-foto

Status

Description: Semantic photo analysis with hierarchical 2D segmentation and structured object detection.

Port: :5700

Venv: venv present /Users/friedrichstr/ai-mac-studio/agents/media/ai-foto/.venv/bin/python

Git: git main — 2026-03-14 chore: initial commit (ai-platform v2 git infrastructure)

File tree

ai-foto/
  data/
    catalogs/
    chains/
    objects.db
  docs/
    api/
    concepts/
    examples/
    schema/
    _quarto.yml
    cli-reference.qmd
    getting-started.qmd
    index.qmd
    styles.css
  fotocases/
    architectural/
    general/
    nature/
    slides/
    README.md
  fotografer/
    design_annotations/
    static/
    templates/
    __init__.py
    __init__.pyc
    server.py
  object_types/
    _index.json
  scripts/
    interpret_foto.py
    migrate_fotocases.py
  src/
    ai_foto/
  AGENTS.md
  ai-foto
  PROJECT.yaml
  pyproject.toml
  README.md
  uv.lock

Project Goaldraft

[DRAFT — edit this] Semantic photo analysis with hierarchical 2D segmentation and object detection Inferred stage: development. Define the specific deliverable, success state, and timeline.

Success criteria:

[DRAFT] Define 3–5 checkable criteria for project completion

Code	`/Users/friedrichstr/ai-mac-studio/agents/media/ai-foto`
Docs	`/Users/friedrichstr/ai-mac-studio/agents/media/ai-foto/docs`
Service port	`5700`

Path & Service Checks

Check	Path	Status	Detail
code	`/Users/friedrichstr/ai-mac-studio/agents/media/ai-foto`	✓ ok	17 items, last_modified=2026-08-24
docs	`/Users/friedrichstr/ai-mac-studio/agents/media/ai-foto/docs`	✓ ok	11 items, last_modified=2025-12-13
service_port	`5700`	? not_running	2026-08-22T08:37:10.497349+00:00

Documentation

Dependencies

Uses:

ai-core import

README

AI-Foto

Semantic photo analysis with hierarchical 2D segmentation and structured object detection.

Overview

AI-Foto determines the semantic content of photos through:

Hierarchical Segmentation: Dividing photos into field sectors (regions) with parent-child relationships
Object Detection: Identifying objects within segments with their properties and substructures
Growing Repository: Building a collection of object types that expands with each analysis

Installation

pip install -e .

Quick Start

Using the Wrapper Script (Recommended for Dev)

The repository includes a wrapper script ai-foto that automatically handles PYTHONPATH and environment detection.

# Make executable
chmod +x ai-foto

# Run commands
./ai-foto --help
./ai-foto analyze path/to/photo.jpg
./ai-foto fotos sync

Standard Installation

pip install -e .

After installation:

# Analyse a single photo
ai-foto analyze path/to/photo.jpg

# Analyse a folder of photos
ai-foto analyze fotocases/slides/

# List known object types
ai-foto objects list

# Add a new object type
ai-foto objects add --category architectural --name "window_frame"

Configuration

Set your Gemini API key:

export GEMINI_API_KEY="your-api-key"

Documentation

The documentation is not published as a site yet.

Photo Categories

The system supports multiple photo categories with category-specific segmentation:

General: background, foreground, subject, context
Architectural: facade, interior, roof, foundation, landscape
Presentation Slides: header, body, footer, diagram, table, bullet_list, image_block

License

MIT License

Recent commits

SHA	Date	Subject
`69b07fc`	2026-03-14	chore: initial commit (ai-platform v2 git infrastructure)