Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Maps are one of the most critical pieces of infrastructure for human society and the global economy.
But there's a problem.
Most of today's maps are built by deploying dedicated vehicles to collect high-quality, street-level imagery. That's expensive, so mapmaking is consolidated among the few companies with the capital to map roads at global scale. It means maps often don't reflect current realities on the ground.
Uneven coverage and freshness - Typical methods of collecting map data are very expensive, so even the best-funded companies struggle to refresh maps at global scale.
Expensive for businesses - Millions of businesses pay to integrate maps into their products. With so few choices for reliable maps, monopoly pricing makes map data unaffordable.
Today's maps don't understand the "why" - When cars start to travel at 10 mph on a road they normally travel at 40 mph, modern maps don't know if this is due to road construction, a minor fender-bender or a severe crash that will take hours to clear. Even when a user reports an issue through an app, they cannot immediately trust the report without imagery.
The documentation in this repository is intended to help people learn about Bee Maps products and learn how to use them. If you can’t find the answers you need, just fill out this form on our homepage, and our sales team will get in touch with you.
Thank you for being a part of the future of maps.
-The Bee Maps team
Bee Maps designed the Map Features API to allow developers and enterprises to easily access Map Features detected using imagery from the Hivemapper Network.
Bee Maps devices are designed to automatically detect objects such as speed limit signs, traffic light signals, turn restriction signs, etc. When detected objects are properly and successfully geo-located, they become map features that are accessible via the Map Features API.
The list of supported Map Features will constantly expand. Below are the currently supported Map Features by country and region.
US and EU
US and EU
US and EU
US and EU
US and EU
US
US
US (California)
US (Texas and California)
US and EU
US (California)
US (California)
US (California)
US (California and Texas)
Below is an example of the data schema for a map feature.
Description:
id – Unique ID for each map feature.
objectClass – The name of the object class, e.g., speed limit or stop sign.
properties – These provide the context and details for the map feature.
lat/lon – The coordinates for the object.
height – The height of the object.
width – The width of the object.
azimuth – The direction in which the object is facing, e.g., the sign is facing northbound traffic.
confidence – The confidence that this object has been successfully identified and geo-located to create or update a map feature.
frameId – The associated image frame(s) that were used to generate this map feature.
firstSeen – The first time Hivemapper has seen and detected this object.
lastSeen – The last time Hivemapper has seen and detected this object.
For a given map feature ID, you can request its history to review changes. Below is an example of a history for a given map feature.
around the world use mapping apps every day. Millions of organizations, from businesses to government agencies, rely on mapping APIs and services to support their operations. With every year that passes, more and more of the world's 1.5 billion vehicles use advanced technology features that demand richer map data than ever before.
Analysts estimate the mapping industry has a market size of, including mapping apps and APIs, geospatial analytics, GIS, and map data collection.
And yet, most roads are driven every day. In the United States alone, 3.26 trillion miles in 2022, or 8.9 billion miles per day. So why can’t ordinary people, and ordinary cars, collect map data?
, launched in 2022, is an open, global, decentralized mapping network that relies on a mix of humans and AI to build a map from the driving we do everyday. Thanks to its vibrant contributor community, Hivemapper millions of kilometers of road-level imagery every week.
, the owner of this page, is a licensed developer in the Hivemapper ecosystem. Bee Maps provides products and services that make it easy for end users such as enterprises, app developers and government agencies to tap into the rich data on the Hivemapper Network.
Fresher maps for autonomous and driving - Today's maps were made for humans. The maps required by autonomous and semi-autonomous vehicles require far fresher data.
Developer documentation for Map Data APIs can be accessed .
You will need an API Key, see for instruction
is a Python based wrapper that allows you to query map features.
Example Python query for Map Features using
Key aspects of data collection and data products generated from map imagery
A list of cities and countries where mapping collection is occurring each quarter
Here are some of the ways that we implement "privacy by design."
Bee Maps will never ask contributors to share personal information such as phone numbers, credit card numbers, or home addresses, except when absolutely necessary to fulfill a shipment. The Hivemapper Network generates random three-word usernames to protect the identity of its contributors, and asks contributors not to share their usernames anywhere to shield their identities.
The first 500 meters and the last 500 meters of a given trip are automatically discarded. A contributor can adjust this up to 1,000 meters. This feature cannot be disabled.
This feature enables a contributor to define specific geographic areas — be it around their home, workplace, or any other location — as zones where their device will stop recording. Whenever the contributor’s device enters a personal privacy zones, it automatically pauses data collection. This eliminates a contributor’s need to manually stop or disable the device as they navigate spaces that they don’t want to map.
Personal Privacy Zones are never shared with Hivemapper servers, and remain confidential only to the contributor.
Network Privacy Zones extends privacy to sensitive areas where everyone’s data collection will be paused. If a location such as a military base, nuclear power plant, or private industrial facility is designated as a Network Privacy Zone, the Hivemapper Network automatically blocks collection within that area.
The Hivemapper Network uses computer-vision algorithms to blur out faces, bodies, vehicles, license plates and faces to protect the privacy of anyone who happens to be captured in mapping images. For example, the image below shows a submission from a Hivemapper contributor that went through privacy blurring. Blurring occurs automatically on the edge on all device models.
The three-word username of the contributor and the time window when the collection occurred.
The street address where the collection occurred and a description of the precise area, including landmarks such as position relative to adjacent streets.
The type of private information that was collected.
Hivemapper Network contributors and Bee Maps customers who receive HONEY require a crypto wallet to be connected to the Hivemapper and Bee Keeper app. Generally, wallets don’t track personally identifiable information and are not necessarily linked to your identity.
The Hivemapper Network proactively works with government regulators to ensure that our privacy standards meet local regulations and are the best in the industry.
Fresh, geolocated street-level imagery from around the globe
Bee Maps designed the Map Image API to allow developers and enterprises to easily access imagery from the Hivemapper Network.
Forward facing (interior or exterior mounted)
1080p
Lane-level positional accuracy
Frames collected and uploaded every 6 to 8 meters — can be dynamically adjusted in certain regions.
Bee Maps is built with data from the , whose mission is to build a useful map for people everywhere to use while protecting their privacy. The personal information and location data of Hivemapper Network contributors and Bee Maps customers is not the product. For this reason, the network has been designed to be private and anonymous, and to uphold the highest standards of data protection and privacy for both contributors and bystanders.
You can find more information on the , a comprehensive portal that includes all the relevant information about privacy at Hivemapper, including:
The current
The location data in the is not associated with a username. Hivemapper doesn’t share usernames or places that have been mapped, so no one will know those locations unless provided. The Hivemapper Network uses various means to verify mapping data location while allowing contributions to remain anonymous.
All Hivemapper devices make it clearly visible to the contributor when they are actively mapping via LED lights and via the companion app. Bee Maps built on the Hivemapper Network is transparent about the data that contributors are collecting and how it is leveraged to build the network and Bee Maps. Furthermore, the Hivemapper Dashcam is an open-source device, and anyone can inspect the that determines what is collected.
Bee Maps and the Hivemapper Network can assist with blurring or deletion of personal information that is inadvertently collected by the network. If you have a request, please fill out the request form in our or contact us at privacy@hivemapper.com. To help Hivemapper identify the relevant materials, please include the following pieces of information in your message:
In certain countries, we work to ensure that bystanders are aware that a device is collecting for a map, device contributors mount a provided by Hivemapper.
Blockchain technology facilitates a new level of encryption, privacy, and anonymity. The Hivemapper Network is built on the blockchain.
Developer documentation for Map Data APIs can be accessed .
You will need an API Key, see for instructions.
is a Python wrapper for the Map Image API, see GitHub .
Example query Python query using
Before using the API, you can quickly evaluate that was collected via front-facing and side-facing devices.
You can purchase Map Credits via self-serve to use the Map Image API.
In additional to regulatory speed limit signs (in US the white signs with black lettering); Bee Maps supports a wide range of speed limit signs, including:
Speed limit for school zones
Speed limit zone start and zone end (EU)
Minimum speed limit zone start and zone end (EU)
Speed limit for construction zones
Speed limit for trucks
Speed limit with time restrictions
Speed limit with minimum speed
Edge based mapping
The Map AI pipeline integrates multiple deep learning models that perform image segmentation, object detection, and geospatial positioning alignment. These models are trained to identify street level elements in varied environmental conditions and from different vehicle-mounted camera perspectives. Once detected, each feature (speed limits, toll prices, etc.) is associated with a precise geographic allowing for accurate placement on the map and relationship to other objects.
Map AI is a fully automated system that processes crowdsourced street-level imagery from the into a structured global street level dynamic map - everything from lanes to road rules to dynamic events like road construction, and police activity.
It uses a combination of computer vision and spatial AI pipeline to detect, classify, and geolocate assets like traffic signs, road markings, lane configurations, gas prices, and more. This system is designed to run on edge devices like the enabling cost effective, scalable, high-frequency map updates without the need for traditional manual mapping.
To continuously improve performance, Map AI outputs are routed through an optional human-in-the-loop system involving . These contributors validate and correct model predictions through structured review tasks, feeding that data back into the training pipeline. This hybrid approach ensures high-quality feature extraction even in edge cases or geographies with sparse data, and it supports rapid model iteration as new types of features are added to the detection set.
We help you spend more time working on what matters most
Bee Maps Automotive Platform is software defined, allowing you to rapidly iterate on your Spatial AI solution instead of spending time and money building operations teams, new hardware or testing fleets.
You don't need to develop AI models unless you want to. Bee Maps brings a world-class engineering team of computer vision and machine learning experts who specialize in extracting valuable real-world information from sensor data. With the Automotive Platform, you can use our standard off-the-shelf models, bring your own models, or work with us to develop entirely new ones — the choice is yours.
You don't need to hire a data labeling team for AI training. One of the core competencies of Bee Maps is our ability to rapidly develop and improve AI models. With our world-class machine learning team, robust data labeling pipeline and hundreds of millions of kilometers of high-quality street-level imagery at our disposal, we can turn around custom computer vision models for you in days -- not months.
You don't need to modify any of your hardware. With our software-defined Bee camera — compatible with any vehicle with a 12V socket — you can start testing Spatial AI capabilities within a few days without touching your hardware stack.
You don't need to deploy testing vehicles. We regularly ship over-the-air updates to tens of thousands of Bee cameras all around the world so you can start collecting real-world data immediately. That means that within hours, you can start seeing how your Spatial AI capabilities work in real-world driving.
You don't need to build a quality assurance team: Spatial AI is powerful, but the technology isn't perfect. With our powerful human-in-the-loop QA pipeline, and tens of thousands of human auditors working on data operations tasks, you can rest assured that you're working with high-quality data.
Monitor locations with the power of the Hivemapper Network
There are three ways to add locations to the map and view street level imagery.
Click on a Road Segment
Then click on View Location
Download example CSV file
The following types of locations are supported:
Address
Point (Coordinate pair)
Polygon
MultiPolygon
The following attributes/properties are supported:
location: the location you want to Scout formatted as an Address, Point, Polygon or MultiPolygon
type: differentiates between the types of locations (Address, Point, Polygon or MultiPolygon), especially when multiple types are included in the same upload
name: preferred identifier for this location
description: additional details about this location
tags: labels for grouping and filtering locations
The minimum requirements for a GeoJSON upload include the location geometry and type as well as the name property. The description and tags are optional.
View monitored locations in a paginated, sortable, and multi-select list.
Rows are highlighted in bold if they contain fresh imagery that has not yet been seen by users or any members of their organization. Multi-select rows to Add Bursts or Remove Locations in bulk. The Search and Tag filter allows enables users to filter through their list. Additionally, users can upload GeoJSON or CSV files using the Upload button. The Burst Status column provides insight into whether locations have an Active Burst, Expired Burst, or None.
View monitored locations in a paginated list. Scroll left and right to view available image sequences from different times and/or dates. Clicking an image sequence will expand the location with all accessible imagery. Fresh indicators advise of locations with fresh imagery with nearby Burst buttons that enable users to Burst individual locations.
Displays a single monitored location in a focused view. The map on the left displays the location of where images were taken. Click the dots on the map to view location specific imagery. Scroll through photos by clicking the left and right arrows on the imagery or by using the left and right arrows on your keyboard. The date and time images were taken are displayed along with the coordinates of where the image was taken can be viewed and copied from the top left corner of the image. Expand the image full screen by clicking the icon on the top right corner of the image.
Displays all monitored locations on a map. Clicking a monitored location will expand the location details. Locations in yellow display locations with active Bursts.
Our roadwork construction detections encompass the following data attributes:
Where the construction impact starts on a road
Where the construction ends on a road
Number of lanes impacted.
We also have an awareness if construction ended if the road was mapped and construction was not detected.
Console is where you can purchase Map Credits, access data, create organizations, and track usage.
Each Map Credit costs $0.02. This means that $1 buys 50 Map Credits, which is roughly equivalent to one kilometer worth of imagery from a single week, or 1 kilometer of a single class of Map Features for a week. There is a minimum purchase of $10, equivalent to 500 Map Credits, with each transaction.
To use Bee Maps APIs, an API key must be generated and active. To generate an API key:
Click Generate API Key
Active API keys will will be listed under Active API Keys
In your API call, add the base64 encoded username:APIKey
After generating an API key and purchasing Map Credits, you now have access to the Map Image API, Map Features API, and Scout.
Scout is perfect for anyone who wants to instantly begin accessing the Hivemapper Network’s fresh, high-quality imagery without directly working with the .
Visit and login using the preferred method: Continue with Google. Alternatively, you can link a Phantom account or use an email address with Breeze.
Credits are needed to view imagery with . Once logged in, navigate to and click Purchase Map Credits to load Map Credits into your account.
If you're a customer interested in accessing the world's freshest map data products, is the place to go. This one-stop shop offers all the core features needed to access Bee Maps products.
Get started by logging into using a Google account. You can also create/link a Phantom or Breeze account, but Google is the preferred login method for Console.
Please do not sign into the website with Google if you are also contributing to the map with a device or completing AI Trainer tasks. Please use Phantom or Breeze instead to receive rewards without interruption. See the for details.
Map Credits are needed to access Bee Maps data products, including the Map Image API, Map Features API and Scout. Map Credits can only be purchased after logging in. Once logged in, you can purchase Map Credits through .
Navigate to
On-Demand, Targeted Coverage
The Hivemapper Network allows developers and their customers to request on-demand and targeted mapping of a specific location using its Bursts functionality. The developer or customer must pay a small fee to fund a bounty for map contributors, who are rewarded for mapping the location.
The before-and-after example below demonstrates how Bursts result in increased coverage of a neighborhood.
Log into Scout, purchase Map Credits, and create locations. Once locations are created, you'll gain the ability to activate a Burst in any given location. This can be done two ways.
Select a location(s) from List View by hovering over the row and selecting the check box. Multi-select capabilities allow you to select multiple locations at once.
Once a location(s) is selected, click the Add Burst button to enable the checkout flow.
Once activated, the Burst Status of the location will update with the expiration date. Locations are unable to be removed from your account while a Burst is active.
In Location Detail view, clicking the Burst button will trigger the Burst activation checkout flow.
Once activated, the Burst button will be replaced with the updated Burst status and expiration date. Locations are unable to be removed from your account while a Burst is active.
Once a customer has created a Burst in Scout, the Hivemapper Network companion app will show the locations to contributors as prioritized bounty areas. If a contributor maps an area with a Burst, updated imagery will appear in Scout within a few hours.
Bursts expire after 7 days. Map Credits will be refunded to the creator of the Burst if no contributor collects imagery in the location during this 7-day timeframe. Once the Burst expires, the functionality to place a Burst in a location is reenabled.
Once you have an active SIM card, it’s time to authenticate:
While the Bee is off, insert the SIM card into the designated SIM card slot
Turn the Bee LTE on
If successful, a pop-up should appear stating “Now mapping with LTE”
Reboot the Bee LTE device
From the Bee Maps App, click on Enable Beekeeper
If not prompted, go to Settings > Bee > Advanced and enable Beekeeper
Enter and verify your email address e.g. bob@email.com
Open the email you received and click on Visit Beekeeper
Login into Beekeeper's web application
Add your fleet name e.g. Bob Logistics
The new Bee LTE device should appear in the Monitoring list view
Once a device is added to Beekeeper, mount it securely on the vehicle’s dashboard or windshield.
Ensure proper alignment with the road and adjust for optimal visibility.
To add devices to your existing fleet, simply repeat step 1 and 2 above using the same email address you used in Step 2.
For example, in step 2, we used bob@email.com to create the Bob Logistics fleet. So, we would use that same email address for future devices to auto-add them to our fleet.
Overview
Beekeeper is an AI application that sees like your drivers—clear, real-time, intuitive. It learns and guides smarter decisions on routes, deliveries, driver behavior, and more. It works seamlessly with the Bee intelligent camera system deployed to your vehicles to provide real-time location intelligence and tracking enhancing operational efficiency and safety.
Who it's for?
Beekeeper AI is designed for teams with commercial vehicles — delivery, logistics, medical, waste management, local services, and more.
There are two parts of the solution:
Beekeeper AI doesn’t just collect data—it sees and understands the world in real-time. Your fleet gets intelligence directly from the road, helping drivers make smarter decisions on fuel, routes, deliveries, and more.
AI Intelligence
Ask questions and get answers
Get clear analytics and automated insights with intuitive charts and graphs.
Live Monitoring
Get live tracking, historic drive paths, and last-known locations, giving you total control and clarity over your fleet's movements.
Trip History
Provides a comprehensive record of all vehicle journeys.
AI-Powered Event Detection:
Leveraging cutting-edge AI, our system detects events like swerving, harsh braking, rapid acceleration, and speeding in real time. This ensures that potential risks are flagged immediately, allowing you to address issues before they escalate.
Third party accessories that Bee customers love:
Connect the (iOS: / Android: ) with the Bee - leave it connected for 5-10 seconds
Connect the (iOS: / Android: ) with the Bee LTE
Open the page on Beekeeper
See here for a installation guide.
The is an intelligent camera device that is mounted in vehicles. It looks like a dashcam, but inside it is far more than a dashcam. There are three core capabilities of the Bee:
Collects and builds a real-time global map for the
An essential pillar of for commercial fleets
A for change detection and running edge based spatial AI
Buy a
for the Bee
Install and for the Bee
See pricing for Beekeeper AI .
The is a professional grade AI powered camera for your vehicles used by tens of thousands of drivers all over the world. It requires an active cellular data plan.
Beekeeper core product capabilities
The best way to learn what Beekeeper can do is to start using it. In the meantime, here’s a quick overview of core product capabilities:
What is Beekeeper Intelligence?
AI tools that give you automated reports (AI Reports) and the ability to ask questions (AI Assitant) about your fleet and the thing they are seeing.
AI Reports
Reports we create automatically that are related to key questions about your fleet
AI Assistant
What % of my drivers had safety related AI Events in the last 7 days?
Have any of my drivers been near 1600 Pennsylvania Ave in the past 24 hrs?
What cities are my drivers doing the most stops in?
What is Live Monitoring?
Status of every vehicle from - driving, idle, offline and more
Last location
See what they see - fetch an image from the Bee to see what the driver is seeing
Lock your devices to your fleet to safeguard your Bee assets
Stats on your drivers
What is Driver Profile?
Driver mileage metrics
LTE usage metrics
LTE carrier information e.g. IMEI
Bee LTE firmware - checkmark means the Bee is on the latest firmware.
What is Trip History?
Trip History provides a comprehensive record of all vehicle journeys, including:
Start and end times for each trip
Stops made with duration
Total distance and trip duration
Location of each stop
Total trips for the day or week, and total mileage.
What are AI Event detections?
When a driver triggers an AI safety event a 5 to 20 second video is uploaded. Currently supported events include:
Speeding over 85 mph (configurable)
Swerving
Harsh braking
Aggressive acceleration
Search your Beekeeper for:
Locations and addresses
Driver Name
Device ID
License Plate
Driver Status
Vehicle Type
Easy to setup your account and add your team members to Beekeeper.
Edge Platform for Custom Workloads
At its core, the Bee was designed with a modular, plug-in architecture, making it fast and easy to deploy workloads. Starting today, we’re unlocking three powerful capabilities for developers.
You can now deploy your own computer vision models directly on the Bee, running alongside the native Hivemapper Map AI stack. This allows you to tailor object detection to your specific use case—whether it’s identifying utility infrastructure like transformers and poles, or industry-specific features like inspection panels, parking meters, or construction signage.
Models are deployed via OTA (over-the-air) updates using Bee Maps’ geo-targeted infrastructure. You can push models to specific devices by region—at the country, state, or metro level—optimizing inference and data coverage based on your operational needs.
The Bee natively uses a single, edge-optimized YOLO model that performs real-time detection and classification on all frames captured by its 12.3MP camera. Running on a 5.1 TOPS-capable System-on-Module (SoM), the Bee is optimized for low-latency inference, even in offline or bandwidth-constrained environments.
How to Deploy a Custom Model
• We collaborate with partners to train objects using their data.
• Your model is deployed to a small subset of Bee devices to validate accuracy and recall
• Once approved, it’s rolled out to your designated geographies via an OTA update to Bee devices.
You can track changes to specific infrastructure objects—like speed limit signs—by uploading a reference dataset in GeoJSON format. Each object should conform to our feature spec, for example:
Bee Maps automatically partitions and distributes these objects to devices most likely to encounter them. As Bees detect and upload fresh observations, your GeoJSON file is updated. You can subscribe to real-time changes via API to track updates as they occur.
All data generated from your custom edge detections and change detection workflows—along with optional supporting imagery—is automatically streamed from the Bee to your Bee Maps developer account. This uses the device’s built-in LTE and WiFi offload channels, with no additional integration work required.
Simply configure your endpoint, and the Bee will begin transmitting structured visual data in real time, ready for ingestion into your systems or workflows.
Contact us (hi@hivemapper.com) to become a Bee developer.
Frequently Asked Questions for Beekeeper AI
Does Beekeeper require the Bee LTE model?
Yes, and it requires an active cellular data plan
How do I know if the firmware of my Bee devices is up to date?
A checkmark under the driver profile means the Bee is running the latest firmware
Firmware updates are automatic OTA (over-the-air) and require no manual work
Can I use my own SIM card in the Bee LTE?
Yes, as long as it’s compatible with the supported networks in your region
SIM card size is nano
Can I connect the Bee to the OBD-2 port?
Does the Bee require professional installation?
No—it takes under 5 minutes to install. Just plug it in, mount it, and you’re good to go
Is there a microphone or speaker on the Bee?
No
Does the Bee have an internal facing camera to spy on drivers?
No—most data doesn’t require an internal camera, and it’s a bit creepy.
Can I lock devices to my Beekeeper fleet?
Yes, just add them to Beekeeper and they will be locked to your fleet.
Is Beekeeper supported in all countries?
No, its no supported in Iran, North Korea, Russia, China, and a few other smaller countries where we are restricted from selling the Bee LTE devices.
What happens if my vehicle was in an accident?
If a Beekeeper AI Event detection was not triggered we store the last 10 minutes of video on the device, so be sure to turn the device off and contact customer support.
What happens if the Bee devices lose LTE signal?
It will appear in the Monitoring list as "Searching" — once it regains LTE it will upload everything it stored on the device when it had no LTE connection.
Can I see the mount rating for my devices?
Yes, under the Assistance tab there is a report that provides mount ratings for each Bee device.
Are Bee devices in a Beekeeper fleet contributing to Hivemapper
Can I access trip history and events in real-time?
Yes, Beekeeper shows live trip activity and logs events as they happen.
Trip history is also stored and searchable.
Can I assign drivers or vehicles to specific Bees?
Yes, Beekeeper lets you tag and label devices for better organization. See the driver profile
Can I remove a Bee device from my fleet?
Yes, by connecting the Bee App to the Bee you can remove it from your fleet
Can I see how much LTE data my fleet is using?
Yes, under the Assistant tab it shows LTE data usage by device and overall
You can also view LTE usage per device in the Driver Profile
How long are video clips from AI Events retained for?
6 months
How frequently can AI Reports be refreshed?
Every 60 minutes
Already have Beekeeper? Access it
Yes, this works.
Yes, all Bee devices contribute to while at the same time performing all of the core functions for your fleet.
Bee - Native Mapping Detections (eg speed signs, traffic lights, etc.)
Your own proprietary dataset for these objects
Your Custom Object Detections
Your own proprietary dataset for these objects
