gisma / metashapetools Goto Github PK

Dear Chris,

I have a set of RGB images that I have processed using GCP using the otho+ tool.

I follow the steps described in the guide (https://github.com/gisma/MetashapeTools) to obtain an optimized and georeferenced sparse cloud with GCPs. However, when I try to process step 4 (Orthoimage-post-GCP) I get the error: Null point coud and, therefore, I cannot generate the orthoimage on the smoothed mesh.

Point cloud refers to the optimized sparse cloud derived from SfM, or a dense point cloud from MVS? I thought I have to use the SfM sparse cloud to intepolate the mesh and calculate the orthoimage by step 4, is this correct?

Thanks :)

Dear Chris,

I used the Ortho+ plugin to process a set of RGB images over a forest using the latest version of Metashape Pro for linux (2.0.1). Specifically, I used the Best Practices workflow using GCPs to georeference the images. Steps 1, 2 and 3 (from image alignment, manual marking of GCPs to optimisation of the sparse cloud) all work fine, however, when I want to run step 4 I get the error "Null point cloud", this means that the system does not recognise or detect the sparse cloud.

Log error
Agisoft Metashape Professional Version: 2.0.1 build 16069 (64 bit)
Platform: Linux
BuildModel: source data = Point cloud, surface type = Height field, face count = High, OOC version, interpolation = Enabled, vertex colors = 1
Finished processing in 0.010364 sec (exit code 0)
Error: Null point cloud

So I tried to install new Python3 dependencies but the error remained.

Then, I uninstalled the Linux version of metashape pro 2.1.0 and installed the previous version 1.8.5 and when I ran step 4 to create the mesh and the orthomosaic it worked.

So, maybe the error derives from some incompatibility of Ortho + with the latest version of metashape pro version 2.0.1 for Linux

Best
Christian