Real-Time Image and Video Anonymization for Edge Devices

Image and Video Blurring SDK for Edge Solutions and Camera Systems.

  • Compliance: Ensure compliance with data protection regulations like GDPR and CCPA.

  • Real-Time: Accelerate data utilization through real-time anonymization, avoid post-processing bottlenecks.

  • Cost- and Time Saving: Avoid expensive and time-consuming post-processing anonymization.

  • Easy Integration: Use our SDK (C++, Python) in your workflows.

  • Flexible Deployment: Runs on AI accelerators, virtual servers, edge devices, on-prem.

Solution Features

Powerful Anonymization, Optimized for Edge

Blurring

Blur faces, persons, vehicles and license plates.

Real-Time Processing

Anonymize images and videos instantly for immediate use.

Flexible SDK

Easily integrate Celantur Edge into your C++ or Python applications.

Model only

Work with a tailor-made ML Model only (ONNX, etc.).

AI Accelerator Support

Deploy on Nvidia GPUs (including Jetson) and other AI Accelerators.

GPU & CPU Support

Leverage GPU or CPU for optimal performance.

Metadata

Extract bounding boxes and segmentation masks.

I/O

video4linux, ffmpeg, jpg/png, any other interface.

Utilities

Profiling built in, Object Tracking for Videos, Tiling Options, etc.

Contact Us. We'll get back to you shortly

FAQ

How does Celantur Edge work?

Celantur Edge comes as an SDK tailor-made for your specific use case and system. It contains all the necessary libraries, interfaces and models to anonymize images and videos.

Will Celantur Edge always be able to deliver real-time image and video anonymization processing?

Celantur Edge's real-time performance depends on hardware capabilities and workload. While it's designed for efficiency, factors like hardware power and data volume can influence processing speed. If required, we'll help you with specifying the optimal hardware specifications.

How much does it cost?

Pricing is customized for each project and includes a core product license fee, as well as optional fees for services (time and material) and ongoing support. Please request a custom quote.

Can I just get a model to use in my existing inference infrastructure?

Yes, we can provide machine learning models in several formats (ONNX, etc.).

How do I get early access?

Please contact us with the form on the top of this page. Alternatively, send an email to hello@celantur.com.

Which types of object classes does it support?

Faces and license plates. Support for persons and vehicles is coming soon.

Testimonials

Trusted by industry-leaders. Here's what they say about Celantur.

"Our customers operate all over the globe and require solutions that meet the data privacy requirements that customers have come to expect in the industry. Celantur has enabled us to roll out a global solution, which we can expand and adapt according to our customers' needs. We are confident that they are the partner with whom we will be able to tackle scenarios we need to meet, no matter how complex."

Georg Schroth

CTO and Co-founder
NavVis GmbH

"With a GDPR inflicted raise of data privacy awareness, anonymization has become a critical task in the processing chain of remote sensing data.
The Celantur solution proved to be a reliable tool for the various kinds of imagery collected with RIEGL laser scanning systems. Even under difficult light conditions the software achieved an impressive detection rate."

DI Christian Sevcik

Manager Strategic Software Alliances
RIEGL Laser Measurement Systems GmbH

Latest Blog Posts about Edge AI

Edge AI Learnings: CPU Architectures, GPU Capabilities, and Challenges with Nvidia Jetson

We delve into the intricacies of CPU choices (ARM vs. x86), unlock the mysteries of GPU capabilities, and tackle the unique challenges of Jetson platforms.


Contact Us