{"name": "Magic Georeferencer", "package_name": "magic_georeferencer", "description": "AI-powered automatic image georeferencing using MatchAnything deep learning model", "about": "Automatically georeference maps, aerial photos, and sketches using the MatchAnything deep learning model. Simply navigate to the approximate location and let AI find the matching features.\r\n.\r\nFeatures:\r\n- Zero manual GCP placement - fully automated\r\n- Cross-modality support (maps to aerial, aerial to maps, sketches to maps)\r\n- Progressive multi-scale refinement for improved accuracy\r\n- GPU acceleration with automatic CPU fallback\r\n- Confidence-based match filtering\r\n.\r\nRequirements:\r\n- QGIS 3.22 or higher (Windows, macOS, Linux)\r\n- Python 3.9+ (bundled with QGIS)\r\n- 8 GB RAM minimum, 16 GB recommended\r\n- ~2 GB disk space for model cache\r\n- Optional: NVIDIA GPU with CUDA for faster processing\r\n.\r\nExternal Dependencies (auto-installed on first run):\r\n- PyTorch >= 2.0.0 and TorchVision >= 0.15.0 (deep learning framework)\r\n- Transformers >= 4.30.0 (HuggingFace model support)\r\n- OpenCV >= 4.8.0 (image processing)\r\n- SciPy >= 1.9.0 (scientific computing)\r\n.\r\nThe plugin automatically downloads the AI model weights (~400 MB) from HuggingFace on first use.\r\n.\r\nPlatform Support: Works on Windows, macOS, and Linux. GPU acceleration requires NVIDIA CUDA-capable GPU (optional, CPU fallback available).\r\n.\r\nLicense: Apache License 2.0", "homepage": "https://github.com/FungoBungaloid/georefio", "repository": "https://github.com/FungoBungaloid/georefio", "tracker": "https://github.com/FungoBungaloid/georefio/issues", "author": "Magic Georeferencer Team", "tags": ["raster", "georeferencing", "machine learning", "deep learning", "automation", "ai", "image matching"], "downloads": 2907, "latest_version": "1.1.0", "versions": [{"version": "1.1.0", "experimental": true, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 2814, "uploaded_by": "fungobungaloid", "upload_datetime": "2025-11-28T05:06:39.900299"}]}