sparkyniner/Netryx-Astra-V2-Geolocation-Tool

The world's most sophisticated street level image geolocation software

What it solves

Netryx Astra V2 identifies the precise geographic location of a photograph by matching it against a database of street-view imagery. It is designed to solve the problem of locating images that lack metadata, such as blurry photos, cropped screenshots, or partial views of buildings, which traditional geolocation methods often fail to identify.

How it works

The tool uses a three-stage pipeline to achieve high-accuracy results:

  1. MegaLoc Retrieval: The system converts your image into a compact visual descriptor and compares it against a database to find the top 500 most likely candidate locations.
  2. MASt3R Matching: It performs dense 3D-aware matching between your photo and the candidate street-view panoramas, allowing it to recognize locations even from small, partial image overlaps.
  3. Spatial Consensus: It clusters the resulting matches geographically to filter out false positives and identify the most trustworthy location, outputting the final GPS coordinates.

Who it’s for

This tool is intended for forensic analysis, investigative journalism, human rights documentation, and OSINT (Open Source Intelligence) research. It is a specialized tool for detailed investigation rather than real-time navigation or tracking.

Highlights

  • Generational Leap in Accuracy: Uses state-of-the-art models (MegaLoc and MASt3R) to outperform previous geolocation methods, especially on difficult, partial, or cropped images.
  • Community Hub: Users can download pre-built city indexes from the community or upload their own, eliminating the need for every user to perform time-intensive indexing.
  • Fully Offline Workflow: Once an index is downloaded, the entire search process runs locally on your machine without requiring an internet connection.
  • Simplified Pipeline: Replaced a complex, nine-stage system with a streamlined three-stage process that is faster and easier to maintain.
  • Hardware Versatility: Runs on macOS (Apple Silicon), Linux (NVIDIA/AMD), and Windows, automatically utilizing GPU acceleration for faster processing.

Related

  • Project
    zju3dv/EfficientLoFTR

    EfficientLoFTR is a semi-dense local feature matching system that provides high-quality image correspondences with speeds comparable to sparse matching methods.

  • Project
    datalab-to/indra

    Indra OCR is a 1 M‑parameter, CPU‑only OCR model that converts PDFs and images to markdown, plain text, or JSON with detailed layout boxes. It processes up to 49 pages / s (≈230 ms per page) and scores 62.1 on the OmniParseBench benchmark, outperforming Tesseract and PaddleOCR in speed and many accuracy metrics. The package provides a CLI, a Python API, and a local web viewer, and supports Latin, Cyrillic, Greek, Devanagari, Bengali, and Chinese OCR, plus any script via PDF text layers. Optional `--math` flag adds LaTeX‑style math extraction. The model is ~18 MB ONNX, licensed under a modified OpenRAIL‑M (free for research/startups), with code under Apache 2.0.

  • Project
    luxonis/depthai

    The depthai library is the software framework for the Luxonis platform, allowing developers to create AI and computer vision applications on Luxonis hardware.

  • Project
    microsoft/Biodiversity

    A collection of open-source AI tools for biodiversity monitoring, providing models for animal detection in camera-trap imagery, bioacoustics, aerial views, and sonar.