from_pascalVOC_detection

from_pascalVOC_detection(input_folder: Path) → Dataset

Load a pascalVOC detection dataset that follows the official structure.

Folder is assumed to contain three sub-folders:

• “Annotations” containing the annotation xml files

• “JPEGImages” containing the images files

• “ImageSets/Main” containing the detection split files

See specifications

See also

from_pascalVOC_generic()

Notes

• This has been tested against PascalVOC2012

• If loading official detection splits, not all images will be assigned a split value.

• For objects with “parts” (like hands for persons), a new object will be created and a body_id column will link to the corresponding root object.

• actions.<value> columns are converted to boolean and included in the booleanized column actions.

• difficult, truncated and occluded columns are converted to boolean.

• The dataset will remove images without split. If you wish to load all images with available annotation, use from_pascalVOC_generic().

Parameters:

input_folder – Folder containing annotations, images, and split folders.

Returns:

Loaded dataset