Deepness QGIS plugin

Robin sits down with Marek Kraft to unpack Deepness, the open-source QGIS plugin that lets you run object-detection, segmentation and regression models with zero command-line setup and no dedicated GPU. Pick a model from the built-in zoo (roads, buildings, crops, wildlife, floods…), click Run, and the results pop up as standard QGIS layers you can style, query and export. We cover why Marek’s team built it, how ONNX keeps everything cross-framework, and the fast-growing community that has already logged 25 000+ downloads. PROJECT LINKS • GitHub – https://github.com/PUTvision/qgis-plugin-deepness • PUT Computer Vision Lab – https://vision.put.poznan.pl/ • Demo of Deepness – https://youtu.be/uMPapvwoPdw 🚀 TIMELINE 00:00 Intro & episode goals 01:10 Marek’s background — robotics → Earth-observation research 02:05 Why Deepness? Pain points that sparked the plugin 03:45 How it works inside QGIS — one-click inference workflow 05:15 Model zoo & ONNX export pipeline (no GPU required) 07:00 Real-world use cases — wildlife counts, flood maps, ship detection 08:45 Community stats: 25 000+ downloads and counting 10:05 Contributing new models & feature requests 11:40 Roadmap — tiling for huge rasters, batch processing, SAR support KEY TAKE-AWAYS • GIS-native UX: run cutting-edge models without leaving QGIS. • CPU-friendly: inference on any modern laptop; GPUs just speed things up. • Plug-and-play models: export from PyTorch/TensorFlow → ONNX → Deepness in minutes. • Everything stays georeferenced: outputs are normal QGIS layers (polygons, masks, rasters). • Open-source momentum: contributions welcome—models, docs, code or testing. If this conversation helps you, please Like, Subscribe, and share your questions below! Bio: Marek Kraft is an assistant professor at the Poznań University of Technology (PUT), where he leads the PUT Computer Vision Lab. The lab focuses on developing intelligent algorithms for extracting meaningful information from images, videos, and signals. This work has applications across diverse fields, including Earth observation, agriculture, and robotics (including space robotics). Kraft's current research involves close-range remote sensing image analysis, specialising in small object detection for environmental monitoring. He also collaborated on European Space Agency projects aimed at extraterrestrial rover navigation and autonomy, making use of his knowledge of embedded systems. His research has led to over 80 publications, several patents, and a history of securing competitive research grants. Kraft is a member of IEEE and ACM.