Skip to main content

Metashape

TOOL
📖5 min read
Latest Version: 2.3.2
Agisoft Metashape - Complete Tutorial (Cloud, Mesh, DSM, DTM, Classify, Orthoimage - No GCPs) by Geospatial Tips

Licensed Product

Metashape is commercial software in three editions. A trial is available; ongoing use requires a license. Record the edition in your method section.

Georeferenced Output

With ground control points it produces georeferenced meshes, orthomosaics, and DEMs that import directly into QGIS in real-world coordinates.

Documented Workspace

The workspace stores the full operation history and can export a report with every setting. That file is your reproducibility record.

Introduction​

Metashape (Agisoft) is the commercial reference for the photogrammetry pipeline. It runs the same stages as the open-source tools (COLMAP, Meshroom): calibration, structure from motion, dense cloud, mesh, texture. It adds a guided ground control workflow, georeferenced orthomosaic and DEM output, and a workspace that records every operation.

Pick Metashape when you need georeferenced outputs for GIS work, a supported product with a documented trial path, or a report export for the case file. Pick the open-source tools when cost, scriptability, or an inspectable algorithm matters more. The method itself is in the photogrammetry technique.

Installation​

  1. Download the installer for your platform (Windows, macOS, Linux) from the official downloads page.
  2. Install the license. A free trial is available for evaluation. For a project, install the license for the edition you purchased (Standard, Professional, or Developer).
  3. Verify the version in Help > About and record it. The current release is 2.3.2 as of September 2026.

Usage​

The pipeline is a sequence of workspace operations. The workspace file stores the order of operations and every parameter.

Create a workspace and load photos· workspace file

Create a new workspace and import the image set. What goes in: photos or extracted frames. What comes out: a workspace file that records the whole run. Name it after the case. If you have ground control points, import them in this step, before alignment, so the calibration uses them.

Calibrate and align the cameras· camera poses, RMS error

Run camera calibration, then track. The application estimates every camera position and the sparse cloud. Check the reprojection error (RMS): low sub-pixel values mean a stable registration. If you imported GCPs, the report shows the error per control point, which is the number a reviewer will ask for.

Build the dense cloud· dense point cloud

Generate the dense point cloud. Filter it if the surface allows: remove outliers on edges where the coverage is thin. The dense cloud is the geometry you will measure against, so check it against known features before you build the mesh.

Build the mesh and texture· mesh, texture, DEM

Build the textured mesh. For site work, also export the DEM and the orthomosaic in this step: both come out georeferenced when the workspace used GCPs.

Export· OBJ, PLY, DEM, report

Export the mesh (OBJ or PLY), the geospatial outputs (DEM, orthomosaic), and the workspace report. What comes out: model files for Blender, GIS layers for QGIS, and the report that documents the run.

In an Investigation​

Use the block below to document how you ran Metashape on the evidence.

Documentation: how to document Metashape

In Your Method Section​

Guidance: Describe the purpose, the tool version, and the input data.

  • Investigative Objective: To reconstruct the 3D geometry of the site from a set of 2D images, georeferenced where control points exist.
  • Software Version: Metashape [Version] [Edition] (e.g., 2.3.2 Professional).
  • Input Data: [Image set name/ID], [N] images derived from [source footage/document], Source Hash: [First 6 digits].

In Your Decision Log​

Guidance: Record the specific procedural parameters and integrity checks.

  • Specific Settings: Calibration mode (auto vs. fixed focal), GCP file name and point count, dense cloud quality setting, any filters applied.
  • Assumptions: The scene was static during capture; camera intrinsics were [auto-calibrated / fixed].
  • Integrity Check: Analysis ran on a working copy, never the original evidence file.
  • Date Performed: [YYYY-MM-DD].

Verification & Mitigation​

Guidance: How did you verify accuracy and mitigate bias?

  • Verification: Read the per-point GCP error from the report. Checked the dense cloud against known feature dimensions.
  • Bias Mitigation: Distributed control points across the scene. Excluded control points with outliers before accepting the alignment.

Example Methodology Statement​

"Site geometry was reconstructed with Metashape 2.3.2 Professional from [N] images of [source] (Hash: A1B2C3). [N] ground control points from [survey/RTK set] were used. Camera alignment produced a mean reprojection error of [X.XX] pixels; the worst control point residual was [X.XX cm]. The mesh and orthomosaic were exported in [CRS] and verified against [reference]."

Into Blender​

Metashape exports model and geospatial formats.

  • Mesh (OBJ or PLY): import with File > Import > Wavefront (.obj) (enable the bundled OBJ add-on) or File > Import > Stanford PLY (.ply). The Photogrammetry Importer add-on places the model in a georeferenced scene.
  • Georeferenced scenes: keep the Blender scene in the same custom CRS and origin as your QGIS baseline, so the model aligns with maps and measurements.

Two caveats before you model against the result:

  • Scale: if the workspace used GCPs in meters, the model is in real-world scale and you can measure directly. Without GCPs the scale is arbitrary, the same as any SfM pipeline: anchor it to a known dimension.
  • Check the orientation of the imported object and rotate it onto the Blender ground plane.

Limitations​

  • It is a licensed product. The license cost matters for small cases, and the trial is time-limited.
  • The algorithms are closed: unlike COLMAP, a third party cannot rerun the same code, only the same settings. The workspace report is the record they get.
  • True georeferencing still depends on the ground control points you supply. Bad GCPs produce a bad map.
  • The same SfM weaknesses apply: reflective, transparent, and featureless surfaces, and moving objects, produce holes and false geometry.

Resources​

  • Photogrammetry, the technique this tool implements, with pitfalls
  • Colmap, the open-source command-line alternative
  • Photogrammetry Importer, the Blender add-on for georeferenced PLY import
  • QGIS, for the geospatial baseline that anchors the georeferenced outputs

Sources​

  1. Agisoft Metashape, official site
  2. Metashape downloads, installers and current version 2.3.2
  3. Metashape tutorials and FAQ, official workflow tutorials including the GCP process