Object detectors
This section contains a description of object detectors that can be used by other SDK libraries or separately for detection objects on video. This category consists of the following libraries:
| Library | Description |
|---|---|
| ObjectDetector (Linux and Windows) | Provides a standard interface as well as defines data structures and rules for different object detectors (motion detectors, change detectors, neural networks, etc.). The ObjectDetector interface class does nothing; it just provides an interface, defines data structures, and provides methods to encode/decode commands and encode/decode params. No external dependencies. |
| Gmd (Linux and Windows) | Motion object detector implementation. Provides automatic detection of moving objects in videos. This library is suitable for various types of cameras (daylight, SWIR, MWIR and LWIR), and it ensures accurate detection of small-size and low-contrast moving objects against complex backgrounds. No external dependencies. |
| Ged (Linux and Windows) | Video change detector implementation. Provides automatic detection of subtle changes in videos. This library is suitable for various types of cameras (daylight, SWIR, MWIR, and LWIR), and it ensures accurate detection of small-size and low-contrast objects against complex backgrounds. No external dependencies. |
| DnnOpenVinoDetector (Linux and Windows) | DNN object detector implementation. Provides automatic detection of objects in videos through the utilization of neural networks based on OpenVINO™ runtime. It is optimized and tested for Ultralytics YOLOv8 neural network model and provides simple interface. External dependencies: OpenCV (version >= 4.5, Apache 2.0 license) and OpenVINO™ (Apache 2.0 license). |
| DnnOpenCvDetector (Linux and Windows) | DNN object detector implementation. Provides automatic detection of objects in videos through the utilization of neural networks based on OpenCV. External dependencies: OpenCV (version >= 4.5, Apache 2.0 license). |