Compare two documents visually
Upload two PDFs or images and see, page by page, where they differ — a pixel heatmap marks exactly what changed, even on scans without a text layer.
How it works
- 1. Upload the original (A) and the new version (B).
- 2. We render both pages and compare them pixel by pixel.
- 3. View the heatmap, the highlighted regions and the percentage changed.
A — original / old
B — new
Without a visual diff
- Text-based diff tools can't read a scan or an image-only PDF at all.
- Adobe Acrobat's compare misfires on different page sizes and re-flowed layouts.
- You can see a difference on screen but can't hand someone the exact pages that moved.
- Small position shifts — a different printer driver — get flagged as if the content changed.
With PDFengine visual diff
- A pixel-by-pixel heatmap showing exactly where, and how much, each page differs.
- Works on scans without a text layer, stamped pages and signature blocks.
- Five view modes — side by side, heatmap, onion skin, swipe and difference.
- Export a diff PDF, PNG or JSON; pages are flagged added, removed or moved.
How it works
Upload two versions
Upload the original (A) and the new version (B).
We compare pixel by pixel
We render both pages and compare them pixel by pixel.
Explore the heatmap
View the heatmap, the highlighted regions and the percentage changed.
When to use a visual diff
For scans, layouts and anything where the appearance is the point.
Scanned & image-only PDFs
When there is no reliable text layer, a pixel diff still shows exactly where two scans differ, page by page.
Design & print proofs
Catch a moved logo, a shifted margin or a changed figure that a text comparison would never notice.
Explain what moved
Five view modes — side by side, heatmap, onion skin, swipe and difference — plus a diff PDF, PNG and JSON to share.
Frequently asked questions
How is a visual diff different from a text compare?
A text compare matches the extracted words and reports insertions and deletions; it's precise when the text layer is clean. A visual diff renders both pages and compares pixels, so it catches layout, font and image changes and works on scans that have no text layer.
Can I compare two scanned PDFs?
Yes — that's exactly what the visual diff is for. It renders each page as an image and highlights where the pixels differ, with a "% changed" per page, so a missing text layer is no obstacle.
What do the view modes do?
Side by side shows the pages next to each other; heatmap colours the changed regions; onion skin overlays both pages; swipe wipes between them; and difference shows only the changed pixels. Each highlights the same changes in a different way.
Is a pixel difference proof that a document was tampered with?
No. A visual diff shows where the rendered pixels differ — a signature block that shifted a couple of millimetres from a different printer driver will show up even though nothing meaningful changed. Read the heatmap as "look here", not as a forensic verdict.
What does it cost?
The visual diff is priced at one credit per paired page, so a short document stays predictable to check. Both files are uploaded to render and compare them, then handled under our retention policy.
Want to know if it's the same document at all?
Use Compare for a layered check of bytes, text, visual and metadata.