ns-render#
Note
Make sure to have FFmpeg installed.
usage: ns-render [-h] {camera-path,interpolate,spiral}
subcommands#
- {camera-path,interpolate,spiral}
Possible choices: camera-path, interpolate, spiral
Sub-commands:#
camera-path#
Render a camera path generated by the viewer or blender add-on.
ns-render camera-path [-h] --load-config PATH [--output-path PATH]
[--image-format {jpeg,png}] [--jpeg-quality INT]
[--downscale-factor FLOAT]
[--eval-num-rays-per-chunk {None}|INT]
[--rendered-output-names [STR [STR ...]]]
[--depth-near-plane {None}|FLOAT]
[--depth-far-plane {None}|FLOAT]
[--colormap-options.colormap
{default,turbo,viridis,magma,inferno,cividis,gray,pca}]
[--colormap-options.normalize {True,False}]
[--colormap-options.colormap-min FLOAT]
[--colormap-options.colormap-max FLOAT]
[--colormap-options.invert {True,False}]
[--camera-path-filename PATH]
[--output-format {images,video}]
arguments#
- --load-config
Path to config YAML file. (required)
- --output-path
Path to output video file. (default: renders/output.mp4)
- --image-format
Possible choices: png, jpeg
Image format (default: jpeg)
- --jpeg-quality
JPEG quality (default: 100)
- --downscale-factor
Scaling factor to apply to the camera image resolution. (default: 1.0)
- --eval-num-rays-per-chunk
Specifies number of rays per chunk during eval. If None, use the value in the config file. (default: None)
- --rendered-output-names
Name of the renderer outputs to use. rgb, depth, etc. concatenates them along y axis (default: rgb)
- --depth-near-plane
Closest depth to consider when using the colormap for depth. If None, use min value. (default: None)
- --depth-far-plane
Furthest depth to consider when using the colormap for depth. If None, use max value. (default: None)
- --camera-path-filename
Filename of the camera path to render. (default: camera_path.json)
- --output-format
Possible choices: video, images
How to save output data. (default: video)
colormap-options arguments#
Colormap options.
- --colormap-options.colormap
Possible choices: turbo, default, gray, magma, inferno, pca, cividis, viridis
The colormap to use (default: default)
- --colormap-options.normalize
Possible choices: True, False
Whether to normalize the input tensor image (default: False)
- --colormap-options.colormap-min
Minimum value for the output colormap (default: 0)
- --colormap-options.colormap-max
Maximum value for the output colormap (default: 1)
- --colormap-options.invert
Possible choices: True, False
Whether to invert the output colormap (default: False)
interpolate#
Render a trajectory that interpolates between training or eval dataset images.
ns-render interpolate [-h] --load-config PATH [--output-path PATH]
[--image-format {jpeg,png}] [--jpeg-quality INT]
[--downscale-factor FLOAT]
[--eval-num-rays-per-chunk {None}|INT]
[--rendered-output-names [STR [STR ...]]]
[--depth-near-plane {None}|FLOAT]
[--depth-far-plane {None}|FLOAT]
[--colormap-options.colormap
{default,turbo,viridis,magma,inferno,cividis,gray,pca}]
[--colormap-options.normalize {True,False}]
[--colormap-options.colormap-min FLOAT]
[--colormap-options.colormap-max FLOAT]
[--colormap-options.invert {True,False}]
[--pose-source {eval,train}] [--interpolation-steps INT]
[--order-poses {True,False}] [--frame-rate INT]
[--output-format {images,video}]
arguments#
- --load-config
Path to config YAML file. (required)
- --output-path
Path to output video file. (default: renders/output.mp4)
- --image-format
Possible choices: png, jpeg
Image format (default: jpeg)
- --jpeg-quality
JPEG quality (default: 100)
- --downscale-factor
Scaling factor to apply to the camera image resolution. (default: 1.0)
- --eval-num-rays-per-chunk
Specifies number of rays per chunk during eval. If None, use the value in the config file. (default: None)
- --rendered-output-names
Name of the renderer outputs to use. rgb, depth, etc. concatenates them along y axis (default: rgb)
- --depth-near-plane
Closest depth to consider when using the colormap for depth. If None, use min value. (default: None)
- --depth-far-plane
Furthest depth to consider when using the colormap for depth. If None, use max value. (default: None)
- --pose-source
Possible choices: eval, train
Pose source to render. (default: eval)
- --interpolation-steps
Number of interpolation steps between eval dataset cameras. (default: 10)
- --order-poses
Possible choices: True, False
Whether to order camera poses by proximity. (default: False)
- --frame-rate
Frame rate of the output video. (default: 24)
- --output-format
Possible choices: video, images
How to save output data. (default: video)
colormap-options arguments#
Colormap options.
- --colormap-options.colormap
Possible choices: turbo, default, gray, magma, inferno, pca, cividis, viridis
The colormap to use (default: default)
- --colormap-options.normalize
Possible choices: True, False
Whether to normalize the input tensor image (default: False)
- --colormap-options.colormap-min
Minimum value for the output colormap (default: 0)
- --colormap-options.colormap-max
Maximum value for the output colormap (default: 1)
- --colormap-options.invert
Possible choices: True, False
Whether to invert the output colormap (default: False)
spiral#
Render a spiral trajectory (often not great).
ns-render spiral [-h] --load-config PATH [--output-path PATH]
[--image-format {jpeg,png}] [--jpeg-quality INT]
[--downscale-factor FLOAT]
[--eval-num-rays-per-chunk {None}|INT]
[--rendered-output-names [STR [STR ...]]]
[--depth-near-plane {None}|FLOAT]
[--depth-far-plane {None}|FLOAT]
[--colormap-options.colormap
{default,turbo,viridis,magma,inferno,cividis,gray,pca}]
[--colormap-options.normalize {True,False}]
[--colormap-options.colormap-min FLOAT]
[--colormap-options.colormap-max FLOAT]
[--colormap-options.invert {True,False}] [--seconds FLOAT]
[--output-format {images,video}] [--frame-rate INT]
[--radius FLOAT]
arguments#
- --load-config
Path to config YAML file. (required)
- --output-path
Path to output video file. (default: renders/output.mp4)
- --image-format
Possible choices: png, jpeg
Image format (default: jpeg)
- --jpeg-quality
JPEG quality (default: 100)
- --downscale-factor
Scaling factor to apply to the camera image resolution. (default: 1.0)
- --eval-num-rays-per-chunk
Specifies number of rays per chunk during eval. If None, use the value in the config file. (default: None)
- --rendered-output-names
Name of the renderer outputs to use. rgb, depth, etc. concatenates them along y axis (default: rgb)
- --depth-near-plane
Closest depth to consider when using the colormap for depth. If None, use min value. (default: None)
- --depth-far-plane
Furthest depth to consider when using the colormap for depth. If None, use max value. (default: None)
- --seconds
How long the video should be. (default: 3.0)
- --output-format
Possible choices: video, images
How to save output data. (default: video)
- --frame-rate
Frame rate of the output video (only for interpolate trajectory). (default: 24)
- --radius
Radius of the spiral. (default: 0.1)
colormap-options arguments#
Colormap options.
- --colormap-options.colormap
Possible choices: turbo, default, gray, magma, inferno, pca, cividis, viridis
The colormap to use (default: default)
- --colormap-options.normalize
Possible choices: True, False
Whether to normalize the input tensor image (default: False)
- --colormap-options.colormap-min
Minimum value for the output colormap (default: 0)
- --colormap-options.colormap-max
Maximum value for the output colormap (default: 1)
- --colormap-options.invert
Possible choices: True, False
Whether to invert the output colormap (default: False)