Behavior Understanding for AV
Gather data for autonomous vehicle datasets by capturing human motions in specific indoor and outdoor environments using sensors and cameras.
Distracted Driver Monitoring
Record contributors in simulated driver seat to identify cues of distracted, drowsy, and alert driving without sleeping for 24 hrs.
Alertness Detection
Contributors annotate video footage of driver’s faces to train models to recognize distracted driving.
Movement Detection for Automotive Security Systems
Track real-life behaviors to train in-car security systems by recording contributors simulating specified actions around the vehicle.
Mapping Application Validation with POI
In-field contributors gather real-time data on geographic locations by physically visiting EV charging stations to validate map accuracy.
Voice Command Response Testing & Correction
Collect data in a simulated in-studio environment by having contributors interact with infotainment and navigation systems to test AI voice command responses.
Translation for In-cabin Voice Assistant
Contributors translate sentences and phrases into 18 different languages to train in-cabin voice assistants.
Object Detection & Tracking for Autonomous Driving
Contributors are trained to work on the customer’s platform and perform rolling point cloud annotation services for autonomous driving.
Object Recognition Optimization
Contributors QA and tag attributes of pre-annotated data, such as counting objects and checking box accuracy in videos and images.
Model Error Mitigation
Contributors tag images, video frames/timestamps with data where customer model underperforms so relevant frames can be selected.

THE state of AI and Machine learning for automotive

An exploration of the automotive industry based on data from the 8th annual State of AI and Machine Learning Report



  • FMVSS and Euro NCAP Regulation compliant AI data
  • Diversity and inclusion as a standard for highly accurate AI
  • Support for complex use cases such as families, twins, equipment, and multi geo-location
  • A trusted partner with experience working in secure environments, extracting PII data, and more

Sourcing, screening, and scheduling participants for automotive data collection is easy with our expert managed services. This includes working with local governments to secure necessary permits.

  • Access a diverse participant pool to enable complex or specific scenarios
  • Field team training and management of secure location setup
  • Onsite training, oversight, compliance and quality collection moderation
  • Onsite crew will rent specified vehicles and install equipment in vehicles for collection

Train in-car virtual assistants to understand speech, initiate phone calls, select radio stations, or play music with our customizable tools and managed services.

  • Providing quality transcription services since 1996
  • Dedicated secure facility transcription services in locations worldwide​ in 170+ languages
  • 50+ full time transcription specialists; many with 5+ years transcription experience
  • 10,000+ experienced transcribers in our global crowd

98% Retention rate among Point Cloud contributors in our dedicated facility

  • 3-week training and certification process with onsite trainers​ and 24-hr shift coverage
  • Have our team work in your platform or provide feedback to help you optimize your tools
  • Support for scaling up or down 200 contributors within 1-2 weeks
  • Experience setting up secure facilities, globally, that meet customer requirements
  • Highly competitive, flexible pricing​ including customer success services

Geolancer, our proprietary POI data collection and verification platform, is designed to mitigate inaccuracy, outdated information, and incomplete records in location data.


Our contributors are equipped with a smartphone running the Geolancer app to add or verify Points-of-Interest ​in their neighborhood


  • Latitude + longitude
  • Address
  • Business Category
  • Opening Hours
  • Pictures and custom meta


  • Remove noise and filter duplicates or incomplete data
  • Automated tools identify user patterns that can negatively impact data quality

More than 100M Appen annotations used for driver monitoring, ADAS systems, autonomous driving, mapping annotations and more!


Automated annotation + human-in-the-loop

  • Facial Points
  • Pixel masks for road data
  • Vehicle bounding boxes
  • Facial and License plate detection and blurring


Semantic segmentations, tiled images, or shapes annotations

  • Ontology supported for up to 1000 unique classes
  • Quality based on IOU, golden sets, multi-round review and more
  • Instance based segmentation

Train ADAS and autonomous driving systems for computer vision and increase safety standards for roadways

  • Video Classification: objects, relevance, appropriateness
  • Video Transcription and Time Stamping: including non-verbal events
  • Object Detection and Tracking: with additional Speed Labeling capabilities to persist your annotations through subsequent frames

Client Success stories


Tier 1 Automotive

Use Cases

Autonomous Vehicles


Securing high-quality data annotation with complex computer vision use cases to train autonomous vehicles.


Vision point cloud obstacle labelling, 3D-2D fusion, lane line labeling, panoramic segmentation, and more. Provided extensive training for point cloud and visual data.


Delivering high-quality data with 98% accuracy and annotated 500K+ 2D-3D Fusion frames/week.


Market-Leading Automotive AI Technology Innovator

Use Cases

Distracted Driver Monitoring


Collecting large volumes of demographically diverse data to detect drowsiness.


Provide data collection of 5,500+ recordings delivered in phases targeting a specific weakness in the model.


Supplemented data for the DMS system continues to mitigate injuries and deaths caused by driver distraction and drowsiness.


Multinational Technology Company

Use Cases

Validation of mapping applications with POI


Identifying outdated electric vehicle charging station data and providing regular re-verification of the charging stations.


Appen compared the data collected against the existing databases. Utilized Geolancers to provide data collection and verification.


Updated 3,738 records:
– 15% of their databases were outdated and updated within 1 month.
– 257 inaccessible or restricted POI identified.


Automotive Software Company

Use Cases

Voice Commands for In-Cabin Assistant


Collecting scripted and spontaneous utterances in 7 different languages to improve natural language processing of an in-cabin VA.


Appen activated our global crowd, securing 300+ demographically diverse speakers to produce over 37,000 utterances.


Test data to validate the model in all 7 languages was successfully obtained and the supplemental data was used to train the model in identified error cohorts.

Why choose Appen?


It’s a high-stakes industry with little margin for error. With more than 15 years’ experience partnering with auto companies, you can trust our industry experts to help you launch with confidence.

Learn More


Nothing gets you across the finish line faster than AI powered by quality data. Our enterprise-ready AI-assisted platform excels in data collection, annotation, and human-in-the-loop evaluation.

Learn More


Our global crowd includes a team of specialized autonomous vehicles AI Training Specialists and contributors who are experienced in providing data for ADAS and infotainment systems.

Learn More
  • Any data type
  • 235+ Languages & Dialects
  • 170+ Countries
  • Public Spaces
  • Secured Facilities
  • Onsite
  • In-cabin
  • In-studio

Website for deploying AI with world class training data