In the last version of eCognition (10.2) you have implemented the deep learning classification algorithm called "deep learning based point cloud classification (prototype)".

In the future, it will be possible to create a deep learning model in order to classify a point cloud?

https://support.ecognition.com/hc/en-us/community/posts/4593762081554/comments/4636136444434

2 comments

  • Keith Peterson

    You can create DL models in eCognition today based on image objects (smallest object size being 1 pxl). You can use these models to classify a point cloud.
    The difference to what we have already implamented is that our internal model was based on the native point cloud and does not require rasterizations of point cloud fields.

  • PoliTO_geomatics lab

    Thank you for your answer Keith.

    I was wondering if it would be possible, in the future, to create a DL model based on native point cloud without creating raster for the classification of the point cloud.

Recent Discussions

Need more help?

Enter Knowledge Base Contact Support