SDKs for Image Processing with ML
SDKs for easy integration into your own capturing and recognition applications.
Don't buy a pig in a poke! Full functional unlimited trial-time before you buy.
Create your own cross-platform applications for object recognition.
The same application works under different operating systems and processor types. All functions of the libraries can be used independently of the platform and automatically adapt to the operating system so that the maximum performance of the corresponding operating system and processor is always used for processing.
When programming the libraries, special attention was paid to performance and memory consumption, so that even a rather slow processor such as a Raspberry Pi 4 still works quickly without much graphics support.
Built from Scratch for Maximum Performance
Introducing a completely reimagined .NET MAUI library, purpose-built to harness the full power of MAUI's latest features. Engineered from the ground up, this library takes full advantage of MAUI’s new rendering architecture and platform-native capabilities—delivering unmatched performance, flexibility, and a seamless developer experience.
Optimized for Android and iOS
On Android, the library leverages the latest Camera2 API and modern renderer architecture, enabling faster initialization, smoother previews, and efficient image handling.
On iOS, we tap into the newest camera and buffer management technologies, ensuring a fluid, high-fidelity experience that feels as fast as native apps.
Smart Resolution Handling
Seamlessly decouple your display resolution from your recognition resolution. Show high-quality previews while optimizing performance for image processing or computer vision tasks—no compromises, just smart design.
Ideal for Real-Time Use Cases
From barcode scanning and license plate recognition to customized ML-powered image analysis, this library is tailor-made for real-time camera use cases on mobile. With finely-tuned buffer management and ultra-low latency, your apps will feel responsive and professional.
While Xamarin.Forms is officially deprecated, we continue to offer a fully supported and functionally equivalent version of our camera library for legacy Xamarin projects.
This version includes the same powerful features as the MAUI library—native camera APIs, resolution separation, and smooth performance—ensuring that existing apps can continue to run reliably without requiring a migration.
- Perfect for legacy apps
- No need to rewrite your project
- Ongoing support for your Xamarin.Forms solutions
The license plate recognition processor recognizes the license plate of various European countries and identifies the corresponding vehicle type. The processor is not limited to a single vehicle/license plate. Depending on the image quality and image resolution, several vehicles/number plates are recognized simultaneously.
State-of-the-art ML (machine learning) is used to enable this reliable recognition. The models we use are constantly being expanded and improved by us in order to recognize even more countries and vehicle types even more accurately.
The recognized plates are validated against the country number formats.
Supported vehicle types:
Supported countries:
A,B,BG,CH,CZ,D,DK,E,EST,F,FIN,FL,GB,GR,H,HR,I,IRL,L,LT,LV,NL,P,PL,RO,S,SK,SLO
The barcode and QR code recognition processor recognizes as many codes as possible within an image and decodes them.
This processor makes it possible to recognize and decode several barcodes and QR codes scattered across an image. The developer can use various parameters to influence the recognition, the results and the display of the results.
Depending on the developer's configuration, the results are displayed (e.g. as an overlay).
To improve the recognition of QR codes, a special binarizer has been developed, which enables QR code recognition under special lighting conditions (Impoved DPM Binarizer).
Supported formats:
Why is it always all or nothing and why can't we expand anything ourselves?
This is exactly the question we asked ourselves when analyzing third-party providers.
Our philosophy is completely different:
Only include what you need and allow other developers to extend it.
We have developed a fully modular plug-in framework that allows recognition plug-ins (so-called processors) to be integrated into the image stream in a modular, simple and efficient way and the results obtained to be returned with the results stream.
This means that only what is actually needed is used. This reduces the size of the resulting application and thus saves memory space on the corresponding device.
With our libraries, the developer can choose whether the processors are to be operated sequentially or in parallel. Both modes have their advantages and disadvantages, but this is the only way to take into account the requirements of the application to be created and conserve the resources of the devices.