This guides goal is to guide the scientists in the workflow for segmenting objects into smaller equal parts and create 3D bounding-box objects surrounding them. The application note includes the steps to import the data and object to be segmented.
This guides goal is to guide the scientists in the workflow for segmenting objects into smaller equal parts and create 3D bounding-box objects surrounding them. The application note includes the steps to import the data and object to be segmented.
Source: The above example data was made available through the courtesy of
Mehwish Anwer, PhD
Team of Professor Cheryl Wellington
University Of British Columbia
In order to run the image registration script, the corresponding python environment should be already installed on the workstation using the Anaconda distribution. Detailed instructions on how to set up this environment are provided in Install Anaconda Python for Vision4D article.
In order to run the script, the objects of interest should first be created and/or imported into the pipeline. We suggest first testing this operator on a small subset of the objects. Running it on the large dataset might take a few hours as well. The script can be downloaded here.
Input_Tag: object tag to use for segmentation.
Input_channel: channel to run the script on.
Width_in_pixels_vertical: Vertical width of the segments in pixels.
Width_in_pixels_horizontal: Horizontal width of the segments in pixels.
Once the script is executed, the segments with 3D bounding boxes will be displayed as objects. By default, they will be assigned random color to differentiate between the different objects.