Getting Started

Getting started with Celantur Edge SDK.

Documentation and Samples on GitHub

For more technical information and samples, please refer to the Celantur SDK documentation on GitHub.

Get a Demo

Reach out to our team to request a demo version.

Anonymizing Panorama Images

Settings:

• ROI: 0.0, 0.2 x 1.0, 0.6

• Tiling: 5 x 1

• Tile Overlap 0.1 x 0.0

Real-time blurring of a dashcam live stream with Celantur Edge SDK on an Nvidia Jetson.

Features

Feature	Category	Description	Available since
Bounding box blurring	Blurring	Blur license plates (rectangular shape) and faces (oval shape).	v1.0
Segmentation blurring	Blurring	Blur segmentation masks	v1.1
License plate and face	Detection		v1.0
Vehicle and person	Detection		v1.1
Setting detection thresholds	Detection	User can set detection threshold	v1.0
Image blurring	Image processing	Blurred image as 3-channel matrices.	v1.0
Read and save JPEG images	Image processing	JPEG images can be loaded and blurred images can be saved as JPEG images.	v1.0
Custom tiling	Detection	Control how many tiles an image is split into before inference.	v1.0
Conserve JPEG metdata	Image processing	JPEG metadata (XMP, EXIF) are copied over in the anonymised image.	v1.0
Detection as structured data	Detection	Detection metadata (type and position of bounding box) can be retrieved.	v1.0
TensorRT GPU inference	Inference	Inference on NVIDIA GPU.	v1.1
ONNX CPU inference	Inference	Inference on Intel and AMD processors.	v1.0
OpenVINO CPU inference	Inference	Inference on Intel (and AMD) processors.
Configure multithreading	Inference	Set number of threads for CPU processing.	v1.0
Video processing (plugin)	Video processing		v1.1
Object tracking	Video processing	under development
Webcam video stream processing	Video processing		v1.1
H.254 (video) en- and decoding	Video processing		v1.1
Docker Image			v1.1