OxygenPDF
pdf-to-svg
convert
vector
cricut

PDF to SVG: How to Get Real Vectors, Not a Picture

RohmanRohman8 min read
PDF to SVG: How to Get Real Vectors, Not a Picture

Open an SVG from a PDF to SVG converter in a text editor and you'll often find one line that matters: an <image> tag followed by a few hundred kilobytes of base64. That file is a PNG of your page with an SVG label on it. It has no paths for a cutting machine to follow, and it goes blurry when you zoom in, which is the opposite of what you wanted from a vector format.

That isn't always wrong, and our own converter does it too in one of its two modes. Sometimes a picture of the page is exactly the job. But it's worth knowing which one you're getting, because the file extension won't tell you.

Quick answer: To convert a PDF to SVG with real vectors, the shapes have to be vectors inside the PDF to begin with. Logos, line art, charts and CAD exports usually are; scans and photos never are. In PDF to SVG, use Pick elements to select vector shapes and download them as an editable SVG with real paths. Use Whole pages when you just need each page as an image in SVG form.

What Is Inside a PDF Page

A PDF page is a list of drawing instructions. Some instructions draw paths: move here, curve there, fill it red. Some draw text with an embedded font. Some place an image, a grid of pixels, at a position on the page.

The first kind converts to SVG cleanly, since SVG paths are the same idea in different syntax. The second kind can come along as text or as outlines, with trade-offs either way. The third kind stays a grid of pixels in any format, SVG included.

So the question "can I convert this PDF to SVG?" really has three answers:

  • A logo, a diagram, a chart from Excel or a drawing exported from CAD or Illustrator is made of paths. You can get real vectors out.
  • A scanned page or a photo is one big image. There are no shapes inside it to extract. Getting paths means tracing: software guessing outlines from the pixels, which is a different tool and a different result.
  • A normal text document is mostly text in fonts, with a few lines and boxes. You can get the boxes as paths. The words are better copied as text, or converted with a design app if you need the letter shapes.

A quick way to check: zoom your PDF viewer to 800% on the shape you want. If its edges stay sharp, it's a vector. If they turn soft or blocky, it's a picture, and no converter will give you more than that.

Two Ways to Convert PDF to SVG

Two ways to turn a PDF page into SVG: whole pages become one embedded picture per page, pick elements turns each vector shape into its own editable path, and text, images and clipping masks are left out

OxygenPDF's PDF to SVG has two modes, because those two outcomes are different jobs.

Whole pages: the page as an image in SVG form

Whole pages renders each page at the resolution you pick (72, 150, 200 or 300 DPI) and saves it as an SVG that holds the rendered page as an embedded PNG. One page gives you one .svg; several pages come as a zip with page-1.svg, page-2.svg and so on. You can limit it to a page range like 1-3, 8.

This is the right mode when something only accepts SVG and you need the page to look exactly like the PDF: a web template, a slide tool, a CMS field. It's honest about what it is, though, so here is the cost. On a four-page text report I tested, the whole PDF was 26 KB. One page rendered at 150 DPI came to about 84 KB as a PNG, and base64 encoding inside the SVG adds a third on top of that. You trade a few kilobytes of drawing instructions for a hundred-odd kilobytes of pixels, per page, and the result is only as sharp as the DPI you chose.

If what you want is an image of the page, a plain PNG from PDF to PNG is smaller and opens in more places.

Pick elements: real vector paths you can edit

Pick elements is the mode most people searching for "PDF to vector" actually need. It reads the drawing instructions of a page and turns every painted path into its own SVG <path>, with its fill, stroke, line width, dashes and transparency. Then it shows you the page as clickable shapes.

  1. Open PDF to SVG, drop in the PDF and switch to Pick elements.
  2. Choose the page. Click a shape to select it; shift-click (or tap, on a phone) to add more. Select all grabs every vector on the page.
  3. Recolor the selection if you like: fill, stroke and opacity.
  4. Copy SVG puts the markup on your clipboard for pasting into code or a design tool, or download it as yourfile-elements.svg.

The exported file is cropped to the shapes you picked, with a little padding, so a logo from the corner of a letterhead comes out as just the logo.

Pick elements leaves text drawn with fonts, embedded images and clipping masks out on purpose, and a shape filled with a gradient or pattern comes out in one flat color. It's a shape extractor, for logos, icons, diagrams and chart lines, and it doesn't pretend to rebuild a whole page.

Two Bugs I Fixed While Writing This

I tested Pick elements properly for this post: drew a red one-inch square in a PDF, extracted it, and checked the SVG. Two things were wrong.

Every shape came out black. The PDF library we render with hands colors over as ready-made hex values like #ff0000. Our extractor still expected three separate numbers, read nonsense, and wrote a color the SVG standard treats as invalid, which browsers draw as black. Red logos, blue chart bars, all of it. That's fixed, with a test that a red square stays red and gray and CMYK colors come through as well.

Shapes landed at 75% of their size. The SVG's width and height were plain numbers measured in PDF points (72 to the inch). A plain number in SVG means CSS pixels, which are 96 to the inch, so a one-inch logo arrived in Inkscape or laser software as three-quarters of an inch. It's the same mismatch Illustrator users have complained about for years. The export now writes sizes in points (width="76pt"), so a one-inch shape is one inch in any app that respects SVG units.

PDF to SVG for Cricut and Laser Cutting

Cutting machines are where the picture-versus-paths difference stops being academic. A cutter follows paths. Give it an embedded PNG and there is nothing to cut.

Cricut Design Space doesn't take PDFs at all. Its upload help page lists JPG, BMP, PNG, GIF and HEIC as basic images and SVG and DXF as vector images, and it says SVG and DXF files can't contain pattern fills or clipping paths. Pick elements output fits those rules by design: it's plain filled and stroked paths, with no clipping masks and no pattern fills.

A workflow that works:

  1. Find the page with the design and confirm it's a vector (the 800% zoom test).
  2. In Pick elements, select only the shapes you want cut. Leave out page borders, crop marks and background boxes, or they become cuts too.
  3. Download the SVG and upload it to Design Space as a vector image, or import it into LightBurn, Glowforge or Inkscape for a laser.
  4. Check the size on the canvas against a known measurement before you cut. Every app has its own import quirks, and checking costs less than a sheet of vinyl.

If the design is a scan or a photo, you need tracing instead. Inkscape's Trace Bitmap and Design Space's own image upload both turn pixels into outlines. Expect to clean up the result.

What About the Text?

Text is the hardest part of any PDF to SVG conversion, and every route has a catch.

Kept as live text, it stays editable and small, but it only looks right if the viewer has the same font. PDFs usually embed only the letters they use, so the font often can't come along. Converted to outlines, it looks exactly right everywhere, but each letter becomes a shape you can't retype. Inkscape offers both when it imports a PDF, and its users report the same trade-off: outlines are faithful and heavy, live text is light and sometimes wrong.

Pick elements skips text entirely. If you need a word as cut lines, the cleanest fix is upstream: convert the text to outlines in the app that made the PDF, export again, and the letters arrive as paths that Pick elements can grab.

Where the File Goes

Logos, client artwork and unreleased designs are the files you least want on someone else's server. Both modes of PDF to SVG run in your browser tab: the page is read and rendered on your machine, and the SVG is built there too. Nothing is uploaded. That's how every OxygenPDF tool works.

Going the other way? SVG to PDF puts SVG files back into a PDF.

Frequently Asked Questions

Is a PDF to SVG conversion lossless?

For vector shapes, nearly: Pick elements copies the path geometry and its colors, line widths and dashes, rounded to a hundredth of a point. For whole pages, no. The page is rendered to pixels at the DPI you choose, so it's exactly as sharp as that resolution and no sharper.

Can I convert a scanned PDF to SVG?

You can wrap a scan in an SVG, but you can't extract vectors from it, because a scan has none. To get paths from a scan you need tracing software such as Inkscape's Trace Bitmap, which guesses outlines from the pixels.

Why is my converted SVG so large?

It's almost certainly a picture of the page encoded as base64 inside the SVG. A page at 150 DPI is around a hundred kilobytes that way. Real vector paths for a logo are usually a few kilobytes.

How do I convert a PDF to SVG for Cricut?

Use Pick elements, select only the shapes you want to cut, download the SVG and upload it in Design Space as a vector image. Design Space can't open PDFs directly and rejects SVGs with clipping paths or pattern fills, and Pick elements writes neither.

Does the SVG keep the original size of the design?

Pick elements writes the SVG's width and height in points, so a shape that measured two inches on the PDF page measures two inches in apps that read SVG units. Some cutting apps still apply their own import scale, so check one known dimension before cutting.

Got a logo trapped in a PDF? Pull it out as real vector paths, in your browser, without uploading the file.

Rohman

Written by

Rohman

Rohman built OxygenPDF's client-side PDF toolkit on pdf.js and pdf-lib, including the PDF to SVG converter and its vector element picker, and writes about what actually happens to a document when you process it in a browser instead of uploading it.

Share this articlePost on XLinkedIn

Stop renting your PDF platform.

All 119+ tools free on web. Desktop Pro is $29 once — every desktop tool, the workspace, and batch processing.

  1. $29now
  2. $79after that

14-day money-back guarantee • Works offline • All platforms

We use analytics to understand how our tools are used and improve the experience. No personal files are ever sent.