Upload a vehicle image and see the license plate, region, make, model, color, and generation.
By submitting, I confirm I am 18 or older and agree to the Privacy Policy and Terms of Use.
Uploaded images are processed in real time and not stored.
How did that go?
The page you used is a preview, not the product.
Read a plate and identify a vehicle from your own image
Sighthound Retriever
for AI-powered ALPR and Advanced Vehicle Recognition
Retriever reads license plates and identifies vehicles by make, model, color, and generation, from live camera streams or uploaded footage, running on your own hardware.
Read the vehicle,
not just the plate
The plate is the obvious thing to read. It's also the thing most likely to fail: a bad angle, distance, dirt, glare, a trailer hitch across the numbers.
Retriever reads the plate when it's there. But it identifies the vehicle either way. Make, model, color, and generation (MMCG) come from the vehicle itself, not from the plate.
So the read doesn't collapse to nothing when the plate won't cooperate. You still know what you're looking at, and you can search for it by what it is, not only by what's written on the back of it.
A plate is one signal,
Retriever gives you more
When a plate reads clean, you have your answer. When it doesn't, the vehicle's own details are what let someone resolve it, a partial plate plus a silver mid-2010s SUV is often enough to know exactly which vehicle you're looking at, without pulling footage.
For cars and trucks, that means make, model, color, and generation alongside the plate. Buses, motorbikes, and other vehicles are detected by type.
It also changes what you can search. Not just the plates you already have on a list, any vehicle, by what it is.
What happens when the plate doesn't read?
Dirt. Rain. A bad angle at the gate. A tow bar across the numbers.
Most of the time the read comes back clean. When it doesn't, something still has to happen, someone checks the feed, looks up a record by hand, or waves the vehicle through and sorts it out later.
Every operator we've talked to has an answer to this. Most of those answers involve a person. The cost isn't how often it happens. It's keeping someone available for a case you'd call rare.
Retriever runs your streams
Live dashboards
Watch active camera streams and detections as they happen
Upload processing
Drop in recorded footage and search what it finds
Regions of interest
Draw the zones that matter on a live stream, ignore the rest
Hotlist matching
Flag plates on a watchlist the moment they’re seen
Search
Find a vehicle by plate, or by make, model, color, and generation
Analytics
See what came through, and when
Runs on your hardware,
Stays in your environment
Retriever installs on your own machines through a Docker Compose stack and runs in the browser. It works offline, with no required cloud connection and no dependency on us being reachable.
Your footage never leaves your environment. Compliance stays under your policies and your controls.
On-Premise
Edge Devices
Docker
Multiple OS
Custom Integration
Cloud
Where teams use Retriever
Parking & Access Control
Support vehicle entry, exit, and access workflows.
Enforcement
Identify overstays, unauthorized vehicles, and lot activity.
Smart Cities & Law Enforcement
Support BOLO, patrol, and situational-awareness workflows.
Tolls & Traffic
Read and organize vehicle activity across traffic workflows.
Fleet & Logistics
Organize vehicle activity across fleet and logistics sites.
Retail & Automotive
Understand vehicle activity at retail and automotive sites.
Customer Insights
Review aggregated vehicle trends and activity patterns.
Sighthound Retriever is powered by ALPR+, our state-of-the-art vehicle recognition technology.
Frequently Asked Questions
For more frequently asked questions about Sighthound Retriever and other Sighthound solutions, check out our Frequently Asked Questions page.
If you do not see the answer you are looking for, contact our experts.
-
For license plates, it provides:
License plate detection
License plate recognition (alphanumeric strings)
License plate region recognition (US states, Canadian provinces/territories, and European Union countries)
For vehicles, it provides:
Vehicle detection
Vehicle recognition (make, model, and generation)
Vehicle color recognition
To learn more about the data provided by Sighthound’s Vehicle Analytics Pipeline parameter definitions, Click Here.
-
Absolutely. Sighthound’s Retriever software utilizes advanced vehicle detection technology along with Make, Model, Color, and Generation (MMCG) recognition capabilities. This means that even if the license plate is not visible, our software can still provide accurate details about the vehicle, enabling comprehensive vehicle identification.
-
While our license plate recognition models can read any alphanumeric plate from around the world, our vehicle recognition models are primarily trained for cars sold in the US, Canada, and EU markets. However, the license plate recognition functionality remains universal, allowing for recognition of plates from any part of the world. Find a complete list of supported countries and regions, here.
-
Although Sighthound’s Retriever can work with almost any IP camera, it is recommended that the appropriate camera be used given the use case. For example, most IP cameras should work fine with an access control use case since the speed of vehicles is relatively low. However, a specialized LPR camera is recommended when dealing with fast-moving vehicles (i.e. highways).
-
The Sighthound team is dedicated to providing support and ensuring your successful installation. You can find installation support information in our Developer Portal or refer to the articles in our Knowledge Base. If you are unable to find the answers you need, please feel free to contact us directly. We are here to assist you.
Become a partner
Sighthound empowers thousands of end users with computer vision solutions.
Learn about our partnership opportunities.