Using vision model for document conversion, clarity about picture and table enrichment

Hi - If I am using the simple pipeline for Vision models as per here:

https://docling-project.github.io/docling/usage/vision_models/

1. How do I ensure the picture enrichment (description), table enrichment, and code enrichment are also happening?
2. Is this included by default when processing a document with the VLM model? If not, are there equivalent option flags as done in the standard PDF parsing pipeline:

from docling.document_converter import DocumentConverter, PdfFormatOption
from docling.datamodel.pipeline_options import PdfPipelineOptions
from docling.datamodel.base_models import InputFormat

pipeline_options = PdfPipelineOptions()
pipeline_options.generate_picture_images = True
pipeline_options.images_scale = 2
pipeline_options.do_picture_classification = True

converter = DocumentConverter(format_options={
InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
})

result = converter.convert("https://arxiv.org/pdf/2501.17887")
doc = result.document

3. If I want to include the full images with the descriptions for embedding later, how is this done? Thanks!

Start with the vision models usage page linked in the issue and trace the simple pipeline used there. Check how picture, table, and code enrichment relate to the standard PDF pipeline options, and document the supported behavior and image-inclusion path. Done means the three questions have clear, tested usage guidance.

Written by the indexing model from the issue text.