Remove Watermark From PDF
PDF watermarks come in two flavours. Some are a real PDF layer you can sometimes delete with a PDF editor; many others — especially in scanned documents or files exported from apps — are burned into the page image, so no editor can 'select' them. WatermarkRemoverAI handles the burned-in case: it treats each page as an image, inpaints the watermark region, and rebuilds the page so the text and layout underneath stay intact. That's the diagonal 'SAMPLE' band, the tool's export logo in the corner, or a repeating tiled stamp across the page.
How to remove a watermark from a PDF
Upload the page or PDF
Add the PDF, or export the affected page to an image first. Pick a page where the watermark position is representative — many documents repeat the same stamp on every page.
Mark the stamp
Brush over the diagonal 'DRAFT'/'CONFIDENTIAL' text, the corner logo, or the tiled pattern. For a full-page tiled watermark, cover the whole repeating grid.
Export the clean page
Download the reconstructed page with the body text and tables preserved, then reassemble your PDF if you split it.
Good to know
Only burned-in / flattened watermarks need this tool — if your PDF watermark is a deletable layer, a PDF editor is faster. Opaque stamps that fully cover text leave clean background, not recovered words.
Frequently Asked Questions
My PDF editor can't select the watermark — why?
Because it isn't a separate layer. In scanned PDFs and many app exports the watermark is flattened into the page image, so there's nothing to click. An inpainting tool rebuilds the pixels instead of trying to delete an object, which is why it works where a PDF editor's 'remove watermark' option greys out.
Will the text under the watermark stay readable?
For a translucent stamp over body text, yes — the AI reconstructs the region and keeps surrounding text sharp. Where a heavy opaque stamp completely covers a word, that word can't be recovered because the original pixels aren't there; the result is a clean background rather than invented text.
Do I have to repeat this for every page?
Each page is processed as its own image, so a multi-page document means marking each page (or batching them on Pro). If the same stamp sits in the same spot on every page, the marking is quick and consistent to repeat.
Can it handle a repeating tiled watermark across the whole page?
Yes. Brush over the entire tiled grid. The larger the covered area the more the AI has to reconstruct, so results are strongest on documents with plenty of plain background between text.
Is removing a PDF watermark allowed?
Remove watermarks only from documents you own or are authorised to edit — your own drafts, licensed templates, or files your organisation produced. Don't use it to strip ownership marks from third-party documents you don't have rights to.
Related Guides
Automate it with the API
Send a file, get a job id back, poll for the clean result. Same AI engine as the web tool, in two requests.
curl -X POST https://api.watermarkremoverai.com/v1/watermark/remove-image/ \
-H "Authorization: YOUR_API_TOKEN" \
-F "files=@photo.jpg" \
-F "auto_detect=true"
# => {"uuid": "..."}
# Poll for the result
curl -X POST https://api.watermarkremoverai.com/v1/watermark/results/ \
-d "uuid=..."
import requests
API = "https://api.watermarkremoverai.com/v1/watermark"
headers = {"Authorization": "YOUR_API_TOKEN"}
with open("photo.jpg", "rb") as f:
job = requests.post(API + "/remove-image/",
headers=headers,
files={"files": f},
data={"auto_detect": "true"}).json()
result = requests.post(API + "/results/",
data={"uuid": job["uuid"]}).json()
print(result)
Prefer a manual mask? Send mask=@mask.png instead of auto_detect, or pass mask_x / mask_y / mask_width / mask_height (add mask_is_percentage=true for percentages). Optional dilation=5 grows the masked area.
Ready to clean it up?
Upload, mark the area, and download the clean result. No account needed to try it free.
Open the Tool