When do you actually need 3D orthomosaics vs standard 2D?
Had an interesting discussion with a client yesterday who insisted on "3D orthomosaics" for a flat commercial site. Got me thinking about when the extra processing effort is actually justified.
For those unfamiliar - a standard 2D orthomosaic projects everything onto a flat plane (or the DSM). A true 3D ortho (sometimes called a "true ortho") corrects for building lean, tall structures, and perspective distortion so that features are shown in their true planimetric position. Its significantly more processing time and requires good oblique imagery.
When 2D is fine (most of the time):
- Flat to rolling terrain with no tall structures
- Construction site progress monitoring
- Agriculture and vegetation mapping
- Stockpile volumes (you're using the DSM anyway)
- Most topo surveys
When 3D actually matters:
- Urban areas with tall buildings (facade lean is real)
- Measuring distances between structures at different heights
- As-built documentation where wall positions matter
- Insurance or legal applications where positional accuracy of vertical features is critical
- Creating basemaps for CAD/GIS where building footprints need to be accurate
Honestly for 90% of survey drone work, a standard 2D ortho processed on the DSM is perfectly adequate. The client who asked for 3D orthos on a parking lot was basically asking for a more expensive product that would look identical to the standard one.
What's everyone else's experience? Am I undervaluing 3D orthos or is this a case of clients asking for things they don't need?