search

openphotogrammetry / meshroomcl

MeshroomCL: An OpenCL implementation of photogrammetry with the Meshroom interface
Other
173 stars 8 forks source link

readme

MeshroomCL

MeshroomCL is an extension of the Meshroom photogrammetry software developed by the AliceVision project. Meshroom and MeshroomCL provide an interface to algorithms for reconstructing 3D models from collections of images. While the original Meshroom required CUDA to achieve accelerated performance, MeshroomCL adds components based on the OpenCL API and can run on a variety of GPU platforms (AMD, Intel, NVIDIA, etc.).

Getting Started

Binary executables for the Windows version of MeshroomCL can be downloaded from https://github.com/openphotogrammetry/meshroomcl/releases. Download the file named MeshroomCL-x.y.z-win64.zip for the most recent version of MeshroomCL. After saving and unzipping the zip file, MeshroomCL can be run via the file Meshroom.exe inside the containing folder.

After running MeshroomCL, follow these basic steps to reconstruct a 3D model from a set of images:

  1. Load your images into MeshroomCL by either dragging the files into the Images pane (upper left) or by using the File → Import Images menu option to select them.
  2. Click the green Start button at the top center of the MeshroomCL window.
  3. You will be prompted to save your project. Select Save and specify a project filename (all project files will be stored in a sub-folder here).
  4. Click the green Start button again. Processing will begin.
  5. You can monitor the progress of the processing pipeline by observing the colored bars at the top of the MeshroomCL window and within each individual node (green="finished", orange="in progress", blue="pending", red="error").
  6. When the StructureFromMotionCL node has finished processing, the computed camera positions and sparse point cloud will be automatically displayed in the 3D Viewer pane.
  7. As each following node finishes processing, you can double-click on that node to view its result in the 3D Viewer pane.
  8. When the last node (TexturingCL) finishes, you can double-click on the node to view the textured 3D model in the 3D Viewer pane, or right-click on the node and select Open Folder to access the 3D model files.

Documentation

MeshroomCL contains context-sensitive documentation for each node, viewable via the Documentation button in the Node pane (lower right). In the node Attributes view of the Node pane, hovering the mouse over each attribute name will display a balloon help message.

MeshroomCL is an extension of the Meshroom software, so much of the original Meshroom documentation still applies to MeshroomCL. The Meshroom documentation site is at https://meshroom-manual.readthedocs.io/.

Several introductory videos for using Meshroom are available on YouTube. A user-contributed video tutorial for MeshroomCL is at https://www.youtube.com/watch?v=vBTr6M7__rQ.

Support

For MeshroomCL technical support, please open an issue on the MeshroomCL GitHub at https://github.com/openphotogrammetry/meshroomcl/issues. We welcome your bug reports, questions, feature requests, and other feedback!

Frequently Asked Questions

Is MeshroomCL available for Linux?

Currently, only Windows binaries are available. However, MeshroomCL can be run on Linux systems via Wine, by doing the following:

  1. Install Wine version 6.5 or later
  2. Override Wine's built-in vcomp140.dll by running the winetricks script with the command winetricks vcrun2019. Alternatively, you can obtain a Microsoft-native version of vcomp140.dll and use winecfg to configure Wine to use it.