Skip to content

Run Time Parameters

Dakota Benjamin edited this page Jan 31, 2017 · 12 revisions

There are the following run time parameters that can be used to adjust the results.

usage: run.py [-h] [--images <string>] [--project-path <string>]
              [--resize-to <integer>] [--start-with <string>]
              [--end-with <string>]
              [--rerun <string> | --rerun-all | --rerun-from <string>]
              [--video <string>] [--slam-config <string>]
              [--force-focal <positive float>] [--force-ccd <positive float>]
              [--min-num-features <integer>] [--matcher-threshold <percent>]
              [--matcher-ratio <float>] [--matcher-neighbors <integer>]
              [--matcher-distance <integer>]
              [--opensfm-processes <positive integer>]
              [--use-opensfm-pointcloud] [--cmvs-maxImages <integer>]
              [--pmvs-level <positive integer>]
              [--pmvs-csize < positive integer>]
              [--pmvs-threshold <float: -1.0 <= x <= 1.0>]
              [--pmvs-wsize <positive integer>]
              [--pmvs-min-images <positive integer>]
              [--pmvs-num-cores <positive integer>]
              [--mesh-size <positive integer>]
              [--mesh-octree-depth <positive integer>]
              [--mesh-samples <float >= 1.0>]
              [--mesh-solver-divide <positive integer>]
              [--texturing-data-term <string>]
              [--texturing-outlier-removal-type <string>]
              [--texturing-skip-visibility-test]
              [--texturing-skip-global-seam-leveling]
              [--texturing-skip-local-seam-leveling]
              [--texturing-skip-hole-filling] [--texturing-keep-unseen-faces]
              [--gcp <path string>] [--use-exif]
              [--orthophoto-resolution <float > 0.0>] [--zip-results]
              [--verbose] [--time]

OpenDroneMap

optional arguments:
  -h, --help            show this help message and exit
  --images <string>, -i <string>
                        Path to input images
  --project-path <string>
                        Path to the project to process
  --resize-to <integer>
                        resizes images by the largest side
  --start-with <string>, -s <string>
                        Can be one of: resize | opensfm | slam | cmvs | pmvs |
                        odm_meshing | mvs_texturing | odm_georeferencing |
                        odm_orthophoto
  --end-with <string>, -e <string>
                        Can be one of:resize | opensfm | slam | cmvs | pmvs |
                        odm_meshing | mvs_texturing | odm_georeferencing |
                        odm_orthophoto
  --rerun <string>, -r <string>
                        Can be one of:resize | opensfm | slam | cmvs | pmvs |
                        odm_meshing | mvs_texturing | odm_georeferencing |
                        odm_orthophoto
  --rerun-all           force rerun of all tasks
  --rerun-from <string>
                        Can be one of:resize | opensfm | slam | cmvs | pmvs |
                        odm_meshing | mvs_texturing | odm_georeferencing |
                        odm_orthophoto
  --video <string>      Path to the video file to process
  --slam-config <string>
                        Path to config file for orb-slam
  --force-focal <positive float>
                        Override the focal length information for the images
  --force-ccd <positive float>
                        Override the ccd width information for the images
  --min-num-features <integer>
                        Minimum number of features to extract per image. More
                        features leads to better results but slower execution.
                        Default: 4000
  --matcher-threshold <percent>
                        Ignore matched keypoints if the two images share less
                        than <float> percent of keypoints. Default: 2.0
  --matcher-ratio <float>
                        Ratio of the distance to the next best matched
                        keypoint. Default: 0.6
  --matcher-neighbors <integer>
                        Number of nearest images to pre-match based on GPS
                        exif data. Set to 0 to skip pre-matching. Neighbors
                        works together with Distance parameter, set both to 0
                        to not use pre-matching. OpenSFM uses both parameters
                        at the same time, Bundler uses only one which has
                        value, prefering the Neighbors parameter. Default: 8
  --matcher-distance <integer>
                        Distance threshold in meters to find pre-matching
                        images based on GPS exif data. Set to 0 to skip pre-
                        matching. Default: 0
  --opensfm-processes <positive integer>
                        The maximum number of processes to use in dense
                        reconstruction. Default: 9
  --use-opensfm-pointcloud
                        Use OpenSfM to compute the point cloud instead of PMVS
  --cmvs-maxImages <integer>
                        The maximum number of images per cluster. Default: 500
  --pmvs-level <positive integer>
                        The level in the image pyramid that is used for the
                        computation. see
                        http://www.di.ens.fr/pmvs/documentation.html for more
                        pmvs documentation. Default: 1
  --pmvs-csize < positive integer>
                        Cell size controls the density of
                        reconstructionsDefault: 2
  --pmvs-threshold <float: -1.0 <= x <= 1.0>
                        A patch reconstruction is accepted as a success and
                        kept if its associated photometric consistency measure
                        is above this threshold. Default: 0.7
  --pmvs-wsize <positive integer>
                        pmvs samples wsize x wsize pixel colors from each
                        image to compute photometric consistency score. For
                        example, when wsize=7, 7x7=49 pixel colors are sampled
                        in each image. Increasing the value leads to more
                        stable reconstructions, but the program becomes
                        slower. Default: 7
  --pmvs-min-images <positive integer>
                        Each 3D point must be visible in at least minImageNum
                        images for being reconstructed. 3 is suggested in
                        general. Default: 3
  --pmvs-num-cores <positive integer>
                        The maximum number of cores to use in dense
                        reconstruction. Default: 9
  --mesh-size <positive integer>
                        The maximum vertex count of the output mesh. Set to 0 for no decimation. Default:
                        100000
  --mesh-octree-depth <positive integer>
                        Oct-tree depth used in the mesh reconstruction,
                        increase to get more vertices, recommended values are
                        8-12. Default: 9
  --mesh-samples <float >= 1.0>
                        Number of points per octree node, recommended and
                        default value: 1.0
  --mesh-solver-divide <positive integer>
                        Oct-tree depth at which the Laplacian equation is
                        solved in the surface reconstruction step. Increasing
                        this value increases computation times slightly but
                        helps reduce memory usage. Default: 9
  --texturing-data-term <string>
                        Data term: [area, gmi]. Default: gmi
  --texturing-outlier-removal-type <string>
                        Type of photometric outlier removal method: [none,
                        gauss_damping, gauss_clamping]. Default: none
  --texturing-skip-visibility-test
                        Skip geometric visibility test. Default: False
  --texturing-skip-global-seam-leveling
                        Skip global seam leveling. Useful for IR data.Default:
                        False
  --texturing-skip-local-seam-leveling
                        Skip local seam blending. Default: False
  --texturing-skip-hole-filling
                        Skip filling of holes in the mesh. Default: False
  --texturing-keep-unseen-faces
                        Keep faces in the mesh that are not seen in any
                        camera. Default: False
  --gcp <path string>   path to the file containing the ground control points
                        used for georeferencing. Default: None. The file needs
                        to be on the following line format: easting northing
                        height pixelrow pixelcol imagename
  --use-exif            Use this tag if you have a gcp_list.txt but want to
                        use the exif geotags instead
  --orthophoto-resolution <float > 0.0>
                        Orthophoto ground resolution in pixels/meterDefault:
                        20.0
  --zip-results         compress the results using gunzip
  --verbose, -v         Print additional messages to the console Default:
                        False
  --time                Generates a benchmark file with runtime info Default:
                        False

More documentation on cmvs and pmvs available here:
http://www.di.ens.fr/cmvs/documentation.html
http://www.di.ens.fr/pmvs/documentation.html