Image & Video Anonymizer — Blur Faces & License Plates (GDPR / LGPD)

Automatically blur, pixelate or redact faces and license plates in photos and videos, in bulk, for GDPR / LGPD privacy compliance. Paste image or video links and get back anonymized files, ready to publish, share or use for AI training. EXIF, GPS and device metadata is always removed.

▶️ Watch the animated video demo

Demo footage: "Roncesvalles Ave crossing at Galley Ave, PXO Type A lights flashing, July 2026" by Qviri, CC BY-SA 4.0, via Wikimedia Commons. Edited: anonymized with this Actor; the demo image and GIF are shared under the same license.

• 🙂 Face blurring: detects frontal and slightly turned faces, including small faces in crowds
• 🚗 License plate blurring: tuned on real Brazilian (Mercosur) road-camera footage, including night/infrared images
• 🎨 Blur, pixelate or black box, with adjustable padding
• 🧹 Strips EXIF/GPS metadata from every output image
• 🔒 Privacy by design: plates are only located, never read or stored, and images are processed inside the Actor and never sent to third-party AI services
• 🎥 Video: every frame anonymized, output as H.264 MP4 that plays everywhere, original audio kept (or removed)
• 📦 Bulk: thousands of images or hours of video per run, with Google Drive and Dropbox links supported
• 💸 Pay per image or per video minute, and failed files are free

Who is this for?

• Real estate and property listings: street photos with neighbours, passers-by and parked cars
• Drone, dashcam, bodycam and CCTV footage: inspections, mapping, fleet, insurance and security video shared with third parties
• AI and computer-vision teams: anonymize datasets before training or sharing them (GDPR/LGPD data minimisation)
• Media, events and research: publish crowd photos without identifiable bystanders
• Marketplaces and classifieds: hide plates on vehicle listings automatically

How to anonymize images

1. Paste image URLs into Image URLs (JPEG, PNG, WebP, BMP, TIFF).
2. Choose what to hide (faces, license plates, or both) and how (blur, pixelate or solid box).
3. Click Start. The Output tab shows the anonymized images with counts of faces and plates hidden.

How video works

Faces and plates are detected several times per second, and boxes are carried between detections so moving people and cars stay covered.

Video mode	Detections per second	Price per started minute	Best for
Standard (default)	5	$0.25	Street, real estate, drone and CCTV footage
Thorough	10	$0.45	Fast motion, dashcams, night footage

Use Max minutes per video to cap cost on long files. The output keeps the source resolution and frame rate.

Pricing

Pay only for what was processed. No subscription; failed files are free.

What	Price	Example
Image anonymized	$0.005 per image	1,000 listing photos = $5
Video, Standard mode	$0.25 per started minute	10-minute drone flight = $2.50
Video, Thorough mode	$0.45 per started minute	10-minute dashcam clip = $4.50

Paid Apify plans get automatic discounts (see the Pricing tab). Use Max minutes per video or the run's maximum cost setting to stay within budget.

Input example

{
  "imageUrls": ["https://example.com/street.jpg", "https://example.com/team.png"],
  "videoUrls": ["https://example.com/dashcam.mp4"],
  "videoQuality": "standard",
  "keepAudio": true,
  "blurFaces": true,
  "blurPlates": true,
  "method": "blur",
  "detectSmallFaces": false,
  "paddingPercent": 20,
  "outputFormat": "jpeg"
}

Output example

{
  "sourceUrl": "https://example.com/street.jpg",
  "outputUrl": "https://api.apify.com/v2/key-value-stores/.../records/anonymized-00001.jpg",
  "width": 1920,
  "height": 1080,
  "facesBlurred": 3,
  "platesBlurred": 2,
  "metadataRemoved": true,
  "faces": [{ "x1": 812, "y1": 240, "x2": 870, "y2": 312, "score": 0.93 }],
  "plates": [{ "x1": 1210, "y1": 760, "x2": 1290, "y2": 790, "score": 0.97 }]
}

Turn off Include detection boxes if you only need the images.

Use it from your code or AI agent

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("adorable_partial/image-anonymizer").call(
    run_input={"imageUrls": ["https://example.com/street.jpg"], "method": "pixelate"}
)
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["outputUrl"], item["facesBlurred"], item["platesBlurred"])

It also works with Make, Zapier, n8n, webhooks and schedules, and as a tool for AI agents through the Apify MCP server.

FAQ

Is the blur reversible? No. Regions are heavily downsampled before blurring, so the original pixels cannot be recovered. For maximum safety, use solid.

Does it read license plate numbers? No. It only finds where plates are. Nothing is read, logged or stored.

Which plates are supported? The plate model was fine-tuned on Brazilian Mercosur and older Brazilian plates from road cameras. It also detects many other rectangular plates, but accuracy on other countries' formats has not been measured yet, so review critical images.

Will it catch every face? It catches most visible faces, including small ones with Detect small / distant faces on. Faces in strong profile, heavily occluded or very blurry may be missed. Review images where a miss would be critical.

Which video formats are supported? MP4, MOV, WebM, MKV, AVI and most other common formats, up to 4 GB per file. The output is always an H.264 MP4.

Can something be missed in video? Plates heavily blurred by fast motion can slip through in single frames, even in Thorough mode. For critical footage, review a sample.

Where are my images processed? Inside the Actor on the Apify platform. Images are not sent to any external AI API.

Related Actors

• Video Optimizer — reduce FPS, resize and compress videos (e.g. before or after anonymizing CCTV footage)
• Video Frame Extractor for AI — turn videos into clean image datasets, then anonymize them here
• Audio Extractor — video to MP3, WAV or FLAC
• Audio & Video to Text (Whisper) — transcripts and subtitles

Support

Missed a face or plate, or need another plate format? Open an issue in the Issues tab with an example and it will be looked at.