Advantages of SentiSight

• Reliable innovative algorithm, that is tolerant to appearance, object scale, rotation and pose.
• Accurate object detection and processing in real-time.
• Webcams or other low cost cameras are suitable for obtaining object images.
• Available as multiplatform SDK that supports multiple programming languages.
• Reasonable prices, flexible licensing and free customer support.

Technology and SDK

• SentiSight object learning and recognition processes. In the learning phase, SentiSight algorithms extract specific object features from a video stream or single image and save them into what is known as an object's model. Manual and automatic object learning are available. No user interaction required for object recognition except providing a video file with the object or pointing a camera to the scene. Read more
• SentiSight algorithm and technology capabilities. SentiSight implements advanced visual-based object learning and recognition technology. It can be used for a wide range of tasks and includes two different algorithms for object recognition. Both algorithms perform simultaneous multiple object recognition, are able to utilize multi-core CPUs and capable of real-time operation. Read more
• SentiSight 3.0 SDK. The SDK allows the development of object recognition systems for Microsoft Windows or Linux platforms, and includes programming tutorials in several programming languages. A Camera Manager library is included for simultaneous capture from multiple cameras. Read more
• System requirements. SentiSight requires a computer with x86 (32-bit) or x86-64 (64-bit) processor that supports SSE2. Best performance is achieved on systems with 64-bit processors that support SSSE3 instruction set. Microsoft Windows and Linux are supported. A webcam or high-res camera may be used for object images capture. Read more
• Technical specifications. SentiSight is able process a single frame in 0.03 seconds during object learning, and perform recognition at speed of 13,000 - 48,000 object models per second when using local features recognition algorithm on 320 x 240 pixels images. Read more
• Reliability and performance tests. SentiSight algoritm was tested with images from ALOI database. SentiSight recognition rate is 98%-99% for objects with well defined internal structure, and 70%-99% for other objects; image processing speed depends on image size, used algorithm and number of threads for matching. Read more
• Download. SentiSight algorithm demo application and SentiSight 30-day SDK Trial are available for downloading. Read more

Licensing, prices and ordering

• Licensing options. To develop a SentiSight-based product, an integrator should obtain SentiSight 3.0 SDK. A license for a SentiSight installation is required for each PC or each server CPU that runs a SentiSight-based application. Single computer licenses and enterprise licenses are available. Read more
• Prices. SentiSight SDK costs € 339. Prices for additional installation licenses depend on quantity. Read more
• Pricing calculator – allows to determine the cost of ordered products and their shipping charges. Open pricing calculator
• Order online – credit cards, PayPal and bank transfers are acceptable. Order SentiSight SDK online

SentiSight related products

• NVeiler Video Filter – a plug-in for VirtualDub video processing application that is able to detect and hide faces in videos; SentiSight algorithm is used for face tracking feature of the plug-in.