Index PDF Elements - Unified Text & Image Embedding with Metadata

View on GitHub

PDFs contain a wealth of textual and visual elements—from narrative passages to figures, charts, and images. In this guide, you'll learn how to set up a complete pipeline that extracts, embeds, and indexes both text and images from PDFs, ensuring every element is fully linked back with original page metadata.

In this example, we split out both text and images, link them back to page metadata, and enable unified semantic search. We’ll use CocoIndex to define the flow, SentenceTransformers for text embeddings, and CLIP for image embeddings — all stored in Qdrant for retrieval.

Incremental Indexing

CocoIndex supports incremental updates out of box. You can add new PDFs or update existing ones without reprocessing the entire dataset. Only new or modified elements are embedded and indexed on each run.

🔍 What It Does

This flow automatically:

• Extracts both page text and images from PDF files
• Filters out images that are too small or duplicates
• Generates standardized thumbnails for images (up to 512×512 pixels)
• Splits and embeds text elements using SentenceTransformers (all-MiniLM-L6-v2)
• Embeds image elements using CLIP (openai/clip-vit-large-patch14)
• Saves all embeddings to Qdrant, along with metadata that traces each embedding back to its source text or image
• Enables unified semantic search across all extracted text and image content

📦 Prerequisite:

Run Qdrant

If you don’t have Qdrant running locally, start it via Docker:

docker run -p 6333:6333 -p 6334:6334 qdrant/qdrant

📁 Input Data

We’ll use a few sample PDFs (board game manuals). Download them into the source_files directory:

./pdf_elements/fetch_manual_urls.sh

Or, feel free to drop in any of your own PDFs.

⚙️ Run the Flow

Install dependencies:

pip install -e .

Then build your index (sets up tables automatically on first run):

cocoindex update --setup main

Or Run in CocoInsight

cocoindex server -ci main

Define the flow

Flow definition

Let’s break down what happens inside the PdfElementsEmbedding flow

@cocoindex.flow_def(name="PdfElementsEmbedding")
def multi_format_indexing_flow(
    flow_builder: cocoindex.FlowBuilder, data_scope: cocoindex.DataScope
) -> None:
    data_scope["documents"] = flow_builder.add_source(
        cocoindex.sources.LocalFile(
            path="source_files", included_patterns=["*.pdf"], binary=True
        )
    )
    text_output = data_scope.add_collector()
    image_output = data_scope.add_collector()

We define the flow, add a source, and add data collectors.

For flow definition, the decorator

@cocoindex.flow_def(name="PdfElementsEmbedding")

marks the function as a CocoIndex flow definition, registering it as part of the data indexing system. When executed via CocoIndex runtime, it orchestrates data ingestion, transformation, and collection.

Process Each Document

Extract PDF documents

We iterate through each document row and run a custom transformation that extracts PDF elements.

with data_scope["documents"].row() as doc:
    doc["pages"] = doc["content"].transform(extract_pdf_elements)

Extract PDF Elements

Define dataclass for structured extraction - we want to extract a list of PdfPage each has page number, text, and list of images.

@dataclass
class PdfImage:
    name: str
    data: bytes

@dataclass
class PdfPage:
    page_number: int
    text: str
    images: list[PdfImage]

Next, define a CocoIndex function called extract_pdf_elements that extracts both text and images from a PDF file, returning them as structured, page-wise data objects.

@cocoindex.op.function()
def extract_pdf_elements(content: bytes) -> list[PdfPage]:
    """
    Extract texts and images from a PDF file.
    """
    reader = PdfReader(io.BytesIO(content))
    result = []
    for i, page in enumerate(reader.pages):
        text = page.extract_text()
        images = []
        for image in page.images:
            img = image.image
            if img is None:
                continue
            # Skip very small images.
            if img.width < 16 or img.height < 16:
                continue
            thumbnail = io.BytesIO()
            img.thumbnail(IMG_THUMBNAIL_SIZE)
            img.save(thumbnail, img.format or "PNG")
            images.append(PdfImage(name=image.name, data=thumbnail.getvalue()))
        result.append(PdfPage(page_number=i + 1, text=text, images=images))
    return result

The extract_pdf_elements function reads a PDF file from bytes and extracts both text and images from each page in a structured way.

• It parses every page to retrieve text content and any embedded images, skipping empty or very small images to avoid noise.

• Each image is resized to a consistent thumbnail size (up to 512×512) and converted into bytes for downstream use.

The result is a clean, per-page data structure (PdfPage) that contains the page number, extracted text, and processed images — making it easy to embed and index PDFs for multimodal search.

Process Each Page

Once we have the pages, we process each page.

1. Chunk the text

This takes each PDF page's text and splits it into smaller, overlapping chunks.

with doc["pages"].row() as page:
    page["chunks"] = page["text"].transform(
        cocoindex.functions.SplitRecursively(
            custom_languages=[
                cocoindex.functions.CustomLanguageSpec(
                    language_name="text",
                    separators_regex=[
                        r"\n(\s*\n)+",
                        r"[\.!\?]\s+",
                        r"\n",
                        r"\s+",
                    ],
                )
            ]
        ),
        language="text",
        chunk_size=600,
        chunk_overlap=100,
    )

2. Process each chunk

Embed and collect the metadata we need. Each chunk includes its embedding, original text, and references to the filename and page where it originated.

with page["chunks"].row() as chunk:
    chunk["embedding"] = chunk["text"].call(embed_text)
    text_output.collect(
        id=cocoindex.GeneratedField.UUID,
        filename=doc["filename"],
        page=page["page_number"],
        text=chunk["text"],
        embedding=chunk["embedding"],
    )

3. Process each image

We use CLIP to embed the image and collect the data, embedding, and metadata filename, page_number.

with page["images"].row() as image:
    image["embedding"] = image["data"].transform(clip_embed_image)
    image_output.collect(
        id=cocoindex.GeneratedField.UUID,
        filename=doc["filename"],
        page=page["page_number"],
        image_data=image["data"],
        embedding=image["embedding"],
    )

When we collect image outputs, we also want to preserve relevant metadata alongside the embeddings.

For each image, we store not only its embedding and raw image data, but also important metadata like the source file name and page number. This metadata ensures you can always identify which document and page an embedded image came from when retrieving or analyzing results.

Export to Qdrant

Finally, we export the collected data to the target store.

text_output.export(
    "text_embeddings",
    cocoindex.targets.Qdrant(
        connection=qdrant_connection,
        collection_name=QDRANT_COLLECTION_TEXT,
    ),
    primary_key_fields=["id"],
)
image_output.export(
    "image_embeddings",
    cocoindex.targets.Qdrant(
        connection=qdrant_connection,
        collection_name=QDRANT_COLLECTION_IMAGE,
    ),
    primary_key_fields=["id"],
)

🧭 Explore with CocoInsight (Free Beta)

Use CocoInsight to visually trace your data lineage and debug the flow.

It connects locally with zero data retention.

Start your local server:

cocoindex server -ci main

Then open the UI 👉 https://cocoindex.io/cocoinsight

💡 Why This Matters

Traditional document search only scratches the surface — it’s text-only and often brittle across document layouts. This flow gives you multimodal recall, meaning you can:

• Search PDFs by text or image similarity
• Retrieve related figures, diagrams, or captions
• Build next-generation retrieval systems across rich content formats

Compare with ColPali Vision Model (OCR Free)

We also have an example for ColPali

To compare two approaches:

Aspect	ColPali Multi-Vector Image Grids	Separate Text and Image Embeddings
Input	Whole page as an image grid (multi-vector patches)	Text extracted via OCR + images processed separately
Embedding	Multi-vector patch-level embeddings preserving spatial context	Independent text and image vectors
Query Matching	Late interaction between text token embeddings and image patches	Separate embedding similarity / fusion
Document Structure Handling	Maintains layout and visual cues implicitly	Layout structure inferred by heuristics
OCR Dependence	Minimal to none; model reads text visually	Heavy dependence on OCR (for scanned PDFs) and text extraction
Use Case Strength	Document-heavy, visual-rich formats	General image and text data, simpler layouts
Complexity	Higher computational cost, more complex storage	Simpler architecture; fewer compute resources needed

ColPali offers superior vision-text integration for complex documents, while separate text and image embeddings are simpler but may lose important context. Choose based on document type and precision needs.

Support us

⭐ Star CocoIndex on GitHub and share with your community if you find it useful!