Segmenting objects in vector graphics images

A system comprising:

• a memory component;
• and
• a processing device coupled to the memory component, the processing device to:
• determine a region of interest in an image that includes an object based on points that identify the object;
• rasterize the region of interest in the image to form a rasterized region;
• generate a mask using an encoder-decoder network, the mask identifying pixels of the object in the rasterized region;
• receive an input to operate in a full containment mode;
• and
• determine, based on a polyline representing an outline of the mask and the input, paths of vector graphics corresponding to the object as visible paths that are contained in an interior of the polyline.

A non-transitory computer-readable medium storing executable instructions, which when executed by a processing device, the processing device performs operations comprising:

• determining a region of interest in an image that includes an object based on points that identify the object;
• rasterizing the region of interest in the image to form a rasterized region;
• generating a mask using an encoder-decoder network, the mask identifying pixels of the object in the rasterized region;
• determining a polyline representing an outline of the mask;
• receiving an input to operate in a full containment mode or a partial containment mode;
• and
• determining paths of vector graphics corresponding to the object as visible paths that are contained in an interior of the polyline in the full containment mode, or as visible paths that intersect an interior of the polyline in the partial containment mode.