A command line toolkit to generate maps, point clouds, 3D models and DEMs from drone, balloon or kite images. 📷
OpenDroneMap Photogrammetry Stack Guide
An engineering guide for running OpenDroneMap pipelines to reconstruct orthophotos, DSM/DTM models, and 3D point clouds.
mapping Reviewed 8/12/2026
Problem definition
How do I configure and run OpenDroneMap (ODM) for automated drone photogrammetry, orthophoto generation, and 3D point cloud reconstruction from aerial geotagged images?
Steps
- Prepare Geotagged Imagery Dataset
Organize geotagged JPEG image files in an images directory. Verify EXIF metadata contains GPS latitude, longitude, and altitude.
- Configure ODM Processing Parameters
Define processing options via command line or task config: --orthophoto-resolution 2.5 --dsm --dtm --feature-quality high.
- Execute ODM Container Run
Launch ODM via Docker container: docker run -ti --rm -v $(pwd)/images:/code/images -v $(pwd)/odm_orthophoto:/code/odm_orthophoto opendronemap/odm.
- Inspect Output GeoTIFF and Point Cloud
Verify generated orthophoto GeoTIFF in QGIS or ODM viewer; check ground sampling distance (GSD) and coordinate reference system (CRS).
Guide
OpenDroneMap Pipeline
OpenDroneMap combines Structure-from-Motion (SfM) and Multi-View Stereo (MVS) algorithms to process drone aerial imagery into geospatial assets.
Judgment criteria
- Generated GeoTIFF orthophoto successfully loads with spatial reference system aligned to input EXIF coordinates.
- Dense point cloud file (LAS/LAZ) generated without camera pose optimization failure.
Common risks
- Insufficient image overlap (<75% frontal, <60% side overlap) causing SfM bundle adjustment collapse.
- Missing camera calibration EXIF data forcing incorrect focal length initialization.