fuse-depth

The fuse-depth tool fuses depth maps from multiple cameras into a single surface.

The default configuration of this tool depends on algorithm implementations in the CUDA and VTK arrows which will only available if the KWIVER_ENABLE_FFMPEG and KWIVER_ENABLE_MVG CMake flags are enabled.

kwiver fuse-depth    [options] input-cameras-dir input-depths-dir

Required arguments:

input-cameras-dir - name of the directory containing the krtd camera files(default: results/krtd)

input-depths-dir - name of the directory to read depth maps from (default: results/depths)

Options are:

-h, --help

Display applet usage information.

-c, --config arg

Configuration file for tool.

-o, --output-config arg

Output a configuration. This may be seeded with a configuration file from -c/–config.

-l, --input-landmarks-file arg

3D sparse features (default: results/landmarks.ply)

-g, --input-local-space-file arg

Input geographic local space file (default: results/local_space.txt)

-m, --output-mesh-file arg

Write out isocontour mesh to file (default: results/mesh.vtp)

-v, --output-volume-file arg

Write out integrated integrated depth data to file (default: results/volume.vti)

-t, --isosurface-threshold arg

isosurface extraction threshold (default: 0.000000).

Default configuration

# Default configuration for the fuse-depth applet

use_gpu = true