Table of Contents | ||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
When using Marso, photogrammetry must only be done with images generated from Marso, rather than using the RAW images, or your own JPEGs/EXRs.
These are generated using the DEVELOP step.
Photogrammetry Images
Once you have the capture photography for a particular subject, you need to create a job for it in Marso.
Create Job
Job Information
Job Name: A descriptive name that will be used to identify the job throughout processing and once the results are ready. e.g Blue_Leather_Handbag
Info |
---|
Letters, numbers, hyphens, and underscores are allowed. Do not use special characters. |
Raw Images Folder: Specify the path to the raw images from the scan to be processed.
Raw Images Extension: Select from the list of supported file formats found in the Raw Images Folder.
Note |
---|
Marso can only ingest camera |
Image Creation
Once you have created a job, the Photogrammetry images need to be created by running the initial DEVELOP step.
Image Customisation
This will return you to the main Marso window, where the selected job is now in the
run queue.
Click start to begin processing all jobs in the queue.
You can pause or stop processing at any time.
Run Develop
Do Solve
You are can now free to follow begin your normal photogrammetry process, as long as the photogrammetry photogrammetry workflow with the images created by Marso are used. You may modify the contents of these images as long as they are not renamed or re-ordered.
See below for specific tips for the photogrammetry solution you use:
Photogrammetry Checklist
Before proceeding with Marso, ensure that you have completed all of the below.
minLevel | 1 |
---|---|
maxLevel | 7 |
location | top |
Camera solve
As well as solving the mesh of the subject, it is important to also get a good solve of the positions of the cameras.
Poorly solved cameras can lead to artifacts in the material measurement later on. We recommend attempting to re-align any bad camera solves or removing those cameras from the 3D solve all together.
Subject Mesh
Poly-Count
A high number of polygons in the mesh may cause Marso to exceed the available VRAM on your system. If this occurs, simplify the mesh and reattempt the process. Marso can produce highly detailed normals, even on medium to low poly meshes. See Examples
Bounding Box
When drawing a bounding box around your subject for mesh reconstruction, it is important to also include any stands or supports. This additional geometry will ensure that pixels are correctly traced onto this geometry and not onto the subject itself.
Using one of the lower cameras (as shown below) can be a helpful reference to see how much geometry is required to avoid unwanted pixels being projected onto the subject.
Good | Bad |
---|---|
Info |
---|
Using a Mask Mesh will allow you to reconstruct just the subject, whilst avoiding occlusion issues by providing a Subject Mesh & Mask Mesh |
UV Unwrap
For Marso to work the subject mesh must be exported with a UV unwrap. By default, a 3D mesh within a photogrammetry application will not have a UV, so you must either run the texturing process within the application, or do a Manual UV Unwrap.
Info |
---|
For optimal UVs we recommend generating the texture at the same resolution or lower resolution than what you wish to output from Marso. Example: A RealityCapture texture of 1024px can be used to create a Marso texture of 1024 -> 8192px , but Marso may create artifacts at 512px. |
Scale
It is essential that the 3D scene is correctly scaled, we suggest using markers to automatically set the scale of a scene within your photogrammetry software.
You can also Re-Scale with Blender
Export
The mesh should be exported as an Alembic (.abc) file, with cameras and normals enabled.
ST Maps
Export the ST Map(s) as exr
so that Marso can account for any lens distortion.
⬅️ Capture
⬅️ Capture