Netryx Street Level Geolocation

---
name: netryx-street-level-geolocation
description: Expertise in using Netryx, the open-source local-first street-level geolocation engine that identifies GPS coordinates from street photos using CosPlace, ALIKED/DISK, and LightGlue.
triggers:
  - geolocate a street photo
  - find GPS coordinates from image
  - street level geolocation
  - netryx geolocation
  - identify location from street view photo
  - osint geolocation tool
  - reverse geolocate image locally
  - build street view index
---

# Netryx Street-Level Geolocation

> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.

Netryx is a locally-hosted geolocation engine that identifies precise GPS coordinates (sub-50m accuracy) from any street-level photograph. It works by crawling street-view panoramas into a searchable index, then using a three-stage computer vision pipeline (global retrieval → geometric verification → refinement) to match your query image against that index. No cloud APIs required — runs entirely on local hardware.

---

## Installation

```bash
git clone https://github.com/sparkyniner/Netryx-OpenSource-Next-Gen-Street-Level-Geolocation.git
cd Netryx-OpenSource-Next-Gen-Street-Level-Geolocation

python3 -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

pip install -r requirements.txt

# Required: LightGlue matching library
pip install git+https://github.com/cvg/LightGlue.git

# Optional: LoFTR for Ultra Mode (difficult/blurry images)
pip install kornia

Platform GPU Support

Optional: Gemini API for AI Coarse Mode

export GEMINI_API_KEY="your_key_here"

Launching the GUI

python test_super.py

macOS blank GUI fix: brew install [email protected] (match your Python version)

Core Workflow

Step 1: Build an Index for a Target Area

In the GUI:

1. Select Create mode
2. Enter center latitude/longitude of target area
3. Set radius (km) and grid resolution (default: 300)
4. Click Create Index

Index is saved incrementally to cosplace_parts/ — safe to interrupt and resume.

Time/size estimates:

Step 2: Search (Geolocate an Image)

In the GUI:

1. Select Search mode
2. Upload a street-level photo
3. Choose method:
   • Manual: Provide approximate center coords + radius
   • AI Coarse: Gemini analyzes visual cues to guess region (requires GEMINI_API_KEY)
4. Click Run Search → Start Full Search