Sighthound Sentry Vision SDK

Accelerating your products with proprietary deeply-learned computer vision technology

Integrating computer vision has never been so simple. The Sighthound Sentry SDK provides a robust and flexible solution that is deeply-learned, platform agnostic, and offers multiple vision intelligence capabilities in one easy-to-use SDK.

Request SDK

Build Once. Deploy Anywhere.

Sentry deploys on platforms ranging from cloud services such as AWS and Azure, to on-premise servers, to embedded systems such as NVIDIA TX1/TX2 boards or devices running iOS or Android.

Complete Computer Vision in One SDK

Embed easy-to-use and robust computer vision into your next product. Avoid multiple conflicting solutions with a single best-of-breed software SDK that enables all major computer vision capabilities, including:

  • • Person & Object Detection
  • • Vehicle Recognition
  • • License Plate Recognition
  • • Face & Emotion Detection
  • • Custom Deep Networks


State-of-the-art deep learning solutions from Sighthound's own computer vision research lab. Patented IP that tops commercial and academic benchmarking tables.

Vehicle Detection & Recognition

Detect vehicles from static or moving cameras, and return the make, model and color of any vehicle sold from 1991 onwards.

Person & Face Detection

Find and track humans as they pass through a scene, find their faces and keep track of the best shot of the face during the scene.

Face Recognition

Improve safety by verifying people are who they say they are. Match attendees to pre-trained galleries for attendance and authentication programs.

License Plate Recognition

Read plates from most countries in the world, reporting the alphanumeric characters and for the US, Canada and the major EU countries, the issuing jurisdiction.

Vehicle Type Identification

Distinguish between buses, trucks, passenger vehicles, and certain corporate fleets, such as UPS, Fedex and USPS delivery vehicles.


Remove personally identifiable information automatically from video feeds or files. Finds faces, license plates and allows other data to be edited out manually.

Sighthound Sentry SDK Capabilities

Combine capabilities to make intelligent vision applications.


Know when something is present.


Know what is present.


Keep track of the number of detections.


Remove or blur identifiable faces or text.


Bring attention to things of interest in an image or video.


Sort like objects together.


Analyze age, gender and facial attributes.


Group images or videos together by person.


Follow specified elements through moving video.

Character Recognition

Read license plates and other text.


Perform a 1:1 match of images of people's faces.


Find makes and models of objects such as cars.


Find known people.


Sense emotion from facial expressions.

Things You Can Do With Sighthound Sentry

Watch various use cases for our SDK.

Public Safety
Suite SDK

Power public safety agencies worldwide

Learn More

Suite SDK

Monitor the interior of vehicles

Learn More

Suite SDK

Bring state-of-the-art deep learning to surveillance

Learn More

Suite SDK

Power your application with intelligent data analysis

Learn More

Integrating the Sighthound Sentry SDK

Our SDK processes various inputs (images, video, RTSP streams, cameras, & custom feeds), and integrates with multiple programming languages to power your application's computer vision capabilities.

Connect With A Vision Expert & Request Sentry SDK

Contact our sales team to learn about integrating Sentry into your product.