So I have a set of photos with bounding boxes already drawn on them. the set is saved as photos on one file and another JSON file with the coordinates of the bounding boxes.
I am running a python recipe with {view_id: choice} to determine how well the bounding boxes are made and what object is contained in the box. The Problem I am running into is that the thickness of the bounding boxes is sometimes blocking the object itself.
To solve this I am trying to do the following: in the return config segment, I am going for "image_manual_stroke_width":2 (instead of the default 4) as shown in the [Available config settings] in the [image_manual] section but with no luck.
I was thinking I could do a {view_id: blocks} and somehow reintegrate the bounding box process adjusting its thickness.
thanks in advance for any input!