Page segmentation of vector graphics documents

A system comprising:

• at least one non-transitory computer-readable medium storing an unstructured vector graphics document and computer-executable instructions for performing a page segmentation process on the unstructured vector graphics document;
• and
• a processing device communicatively coupled to the at least one non-transitory computer-readable medium for executing the computer-executable instructions, wherein executing the computer-executable instructions configures the processing device to perform operations comprising:
• receiving as input from the at least one non-transitory computer-readable medium the unstructured vector graphics document, the unstructured vector graphics document comprising drawing commands;
• generating an element proposal for each of a plurality of areas on a page of the unstructured vector graphics document tentatively identified as being page elements, each of the element proposals generated at least in part based on an analysis of the drawing commands;
• classifying each of the element proposals into one of a plurality of defined types of categories of page elements, the classification generated at least in part based on the drawing commands;
• and
• selecting a set of final classified element proposals within the page of the unstructured vector document by resolving overlapping element proposals within the page.

A computer-readable storage medium storing non-transitory computer-executable program instructions, wherein when executed by a processing device the program instructions cause the processing device to perform a method comprising:

• a step for receiving input comprising an unstructured vector graphics document, the unstructured vector graphics document comprising drawing commands;
• a step for executing a first machine learning algorithm trained to generate an element proposal for each of a plurality of areas on a page of the unstructured vector graphics document tentatively identified as being page elements, each of the element proposals generated at least in part based on an analysis of the drawing commands;
• and
• a step for executing a second machine learning algorithm trained to classify each of the element proposals into one of a plurality of defined types of categories of page elements, the classification generated at least in part based on the drawing commands.