Paper Figures to PowerPoint
Extract all figures, tables, and equations from a research paper PDF and generate a polished PowerPoint deck automatically.
Your lab meeting is tomorrow and you need to present a paper. You're manually screenshotting figures from a PDF, pasting them into PowerPoint, and re-typing table data. For a 12-figure paper, this takes an hour before you've even started thinking about what to say.
Who it's for: graduate students presenting journal club papers, postdocs preparing lab meeting presentations from recent publications, professors creating lecture slides from research papers, research teams doing weekly paper review presentations, scientists presenting literature at conferences
Example
"Turn this Nature paper into a presentation deck" → PowerPoint with all 8 figures extracted at full resolution, 3 tables reformatted for slides, key equations highlighted, figure captions as speaker notes, and a summary slide of main findings
CLAUDE.md Template
New here? 3-minute setup guide → | Already set up? Copy the template below.
npx ccplaybooks add paper-figures-pptx# Paper Figures PowerPoint Generator
Generate a PowerPoint presentation from a Marker-converted paper folder or directly from raw PDFs. The deck includes a title slide with paper metadata followed by one slide per figure, table, and equation, preserving markdown document order for content slides.
## When to Use
Activate when the user wants to:
- Create a PowerPoint from a research paper's figures and tables
- Generate a visual summary deck from a Marker-converted PDF
- Extract all figures/tables/equations from a paper into slides
- Generate PPTX decks from existing Marker output folders
- Generate PPTX decks directly from PDF(s) when markdown does not exist yetGet new playbooks like this one
One email a week with new Claude Code workflows. Free, like everything here.
No spam. Unsubscribe anytime.
What This Does
Converts a research paper PDF into a PowerPoint presentation with one slide per figure, table, and equation. Uses Marker for PDF-to-markdown conversion, then parses the output to build a polished 16:9 deck with a title slide (metadata, DOI, abstract) followed by content slides preserving document order.
Quick Start
Step 1: Install Dependencies
pip install python-pptx Pillow matplotlib
You also need marker_single on your PATH for PDF conversion. See Marker installation.
Step 2: Download the Template
Click Download above to get the CLAUDE.md file, then place it in your project folder along with the scripts from the source repository.
Step 3: Generate a Deck
claude
Say: "Create a PowerPoint from this paper's figures" and provide your PDF path.
Slide Layout
| Slide Type | Content |
|---|---|
| Title | Paper title, authors, APA citation, DOI, truncated abstract |
| Figure | Figure image with heading (optional caption in experimental mode) |
| Table | Table title + native PowerPoint table with header row styling |
| Equation | Label + rendered LaTeX image in highlighted box + context paragraph |
Two Ways to Run
From raw PDF (recommended):
bash scripts/marker_output_to_pptx.sh \
--pdf-input /path/to/paper.pdf \
--marker-output /path/to/marker_output \
--output-dir /path/to/pptx_output
From existing Marker output:
bash scripts/marker_output_to_pptx.sh \
--marker-output /path/to/marker_output
Key Options
| Flag | What It Does |
|---|---|
--all |
Generate decks for all papers in the output folder |
--paper <name> |
Choose one specific paper |
--with-captions |
Experimental: add extracted caption text to figure slides |
--skip-tables |
Omit table slides |
--skip-equations |
Omit equation slides |
--marker-force |
Reprocess previously converted PDFs |
--marker-use-llm |
Use LLM-enhanced extraction for better captions |
Tips
- Caption quality varies — the default figure-only mode is more reliable; use
--with-captionsexperimentally - Reprocess noisy PDFs — use
--marker-use-llmwith--force_ocrfor scanned papers with poor text extraction - Check before generating — verify figure count with
find "$OUT" -type f | rg '_page_[0-9]+_Figure_[0-9]+' | wc -l - Customize the look — edit
generate_paper_pptx.pyconstants for colors, fonts, and slide dimensions
Example Prompts
"Create a PowerPoint from this paper's figures"
"Extract all figures and tables from my PDF into slides"
"Generate a presentation deck from my Marker output folder"
"Make a figure deck from this paper, skip equations"
"Reprocess this PDF with better OCR and generate slides"