• Feature tracking between image pairs using template matching. [example]
• Full processing line from feature tracking to georectification. [example]
• Distorted camera model. This allows the use of cheaper camera setups.
• Avoiding traditional image registration as a pre-processing step as it degrades the images due to resampling. ImGRAFT will instead optimize camera view for each image. [example]
• Projecting between pixel and real world coordinates. 2D ↔ 3D [documentation]
• Easily scriptable as it is a toolbox. [tips]
• Minimal dependencies. (No other toolboxes required).
• Inputs to feature tracking: two images
• Inputs to georectification: a DEM and some ground control points.

We hope you will find this package useful. We would be grateful for any feedback and example use-cases. We are also open to collaboration.