The TinyCD model is both lightweight and effective, able to achieve performances comparable or even superior to the current state of the art models with 13-150 times fewer parameters.
Pre-trained-models
- ESRI has released 10 new pre-trained deep learning models to Living Atlas. These models are available as deep learning packages (DLPKs) that can be used with different ESRI products.
- Determine the location of a given street view image by matching it against a collection of geo-tagged satellite images. Read more about it here.
- Check here for a deep learning method for building height estimation using high-resolution multi-view imagery over urban areas.
- Check out this new method and impressive results of unsupervised semantic segmentation.
- Read here about PEARL - on open sourced platform that allows you to do fast AI-based land classification without writing a line of code.
- Learn how to fine-tune Esri's deep learning models (available through ArcGIS Living Atlas of the World) and refine them with your own training data to improve accuracy for your area of interest.
