General Parameters
Use bash celantur --help to print all the parameters and exit.
Anonymisation
-a OBJECT
--anonymise OBJECT
REQUIRED.
Select what you want to anonymise.
Options: face, license-plate, person, vehicle
--method METHOD
Choose anonymisation method.
Options:
blur: (default) non-reversible blur.blacken: remove colour value from detection.pixelate: low-res rendering of detection.detect: no anonymation.
--bbox
Anonymise whole bounding boxes instead of segmentation.
--debug
Visualise bounding boxes, anonymisation and tiling.
--score
--confidence
Print the confidence scores for the detected objects on images (only if --debug is enabled).
--save-mask TYPE
Save binary mask or instance mask or both (all).
See Segmentation Masks and Metadata.
Options: binary, instance, all
--mask-scale [0..100]
Downscale binary or segmentation mask (only if --save-mask).
--overwrite
Overwrite existing files in output folder. If not specified, files in input folder will be skipped if file with same name exists in output folder.
--quality [0..100]
Quality of output JPEG image.
-f FORMAT
--format FORMAT
See Format / Tiling
--file-type {image,video}
-t {image,video}
Select file type for processing. Default: only images selected. Since v26.01.1.
Blurring gradients
For details see Customize Blurring
--face-anonymization-gradient-start
Margin of gradual face blurring
--face-anonymization-gradient-stop
Padding of gradual face blurring
--license-plate-anonymization-gradient-start
Margin of gradual license plate blurring
--license-plate-anonymization-gradient-stop
Padding of gradual license plate blurring
--kernel-size-face
Kernel size for face blur.
--kernel-size-person
Kernel size for person blur.
--kernel-size-license-plate
Kernel size for license plate blur.
--kernel-size-vehicle
Kernel size for vehicle blur.
Inference
-e INFERENCE_ENGINE
--inference-engine INFERENCE_ENGINE
--model INFERENCE_ENGINE
Select inference model. Current default is segmentatio-omnium
--face-threshold [0..1]
Threshold for face detection.
--license-plate-threshold [0..1]
Threshold for license plate detection.
--vehicle-threshold [0..1]
Threshold for vehicle detection.
--person-threshold [0..1]
Threshold for person detection.
--cpu-mode
Disable GPU support, Using CPU only.
Video processing
--video
Anonymise videos, skip images. Alias for -t video and no -t image
--no-object-tracking
Disables Object Tracking which reduces flickering of anonymized objects in videos.
--keep-bit-rate
Using this flag will encode the anonymised video with a bitrate close to the original video and ensure a nearly identically file size of the processed video, at the cost of 10% to 20% higher processing time.
--no-ffmpeg-merge
Usually after processing the frames, in the post-processing process,
audio and data streams of the input video is merged into the output video, and
the bitrate of the output video is set to be close to the input video if
--keep-bit-rate.
With --no-ffmpeg-merge, you disable this step.
Since v26.01.1.
Format / Tiling
By specifying image format settings with --format FORMAT , -f FORMAT, Celantur Container can achieve improved anonymization results and apply anonymization only to a specific region of an image.
Don't use spaces in the format string. Otherwise Bash has problems interpreting it as one argument.
You can use the option --format to choose the resolution of the input images, eg. --format pano:8000 for an image resolution of 8000x4000.
Use --format whole (default) for all formats not listed below, or if the input images have different resolutions.
pano:4096
4096x2048
pano:5400
5400x2700
pano:5640
5640x2816
pano:7060
7060x3530
pano:7680
7680x3840
pano:8000
8000x4000
pano:7680
7680x3840
pano:8000
8000x4000
pano:8192
8192x4096
pano:11000
11000x5500
whole
default, valid for all resolutions
Ensure that the image resolution matches exactly the predefined resolution in the parameter to avoid processing errors.
Custom tiling for improved results
In certain cases, its beneficial to process high-resolution imagery in tiles, instead of the whole image. Small, distant objects are more likely to be detected.
nx
Number of horizontal tiles
ny
Number of vertical tiles
ox
Horizontal overlap between tiles in pixels
oy
Vertical overlap between tiles in pixels
Limit anonymization to a section of the image
Apply anonymization only to a specified rectangular section of the image. This can result in faster processing time.
number and overlap attribute are required when specifying the section, as seen in the example above.
x1, y1
Top left coordinate of the section
x2, y2
Bottom right coordinate of the section
Configuration (under the hood)
-l FILE
--license FILE
Path of the license file.
-i PATH
--input PATH
--input-dir PATH
Input file or directory.
-o PATH
--output PATH
--output-dir PATH
Output directory.
--log-level LEVEL
Logging level.
Options: DEBUG, DETAIL, INFO, WARNING, ERROR, CRITICAL
--log-dir FOLDER
Location of the log files.
--pid-file FILE
Path of PID file.
Last updated