Try for free
PRICING
CVAT OnlineCVAT Enterprise
COMMUNITY
DiscordLinkedinYoutubeGitHub
CONTACT US
Contact us

#1 Data Annotation Platform for AI and Machine Learning Teams

Transform raw, unstructured images, videos, and 3D data into high-quality, AI-ready visual datasets.
Contact us
About CVAT
CVAT (Computer Vision Annotation Tool) is a data annotation platform for computer vision teams. Originally developed as an internal tool at Intel and released as open-source in 2018, CVAT is now independently operated by CVAT.ai as a robust solution for industry use.

It uniquely combines open-source transparency with enterprise-level scalability, strict data privacy controls, and AI-powered automation, making it ideal for creating high-quality visual AI datasets.

Core Vision Tasks

Image classification
Semantic segmentation
Object detection
Keypoint detection / pose estimation
Instance segmentation

Core Vision Tasks

Image classification
Semantic segmentation
Object detection
Keypoint detection / pose estimation
Instance segmentation

Core Vision Tasks

Image classification
Semantic segmentation
Object detection
Keypoint detection / pose estimation
Instance segmentation

Core Vision Tasks

Image classification
Semantic segmentation
Object detection
Keypoint detection / pose estimation
Instance segmentation

Core Vision Tasks

Image classification
Semantic segmentation
Object detection
Keypoint detection / pose estimation
Instance segmentation

Video-Specific Tasks

Object tracking
Manual, interpolated, and automated labeling with human review
Action recognition
Event tagging
Time-based behaviors, activities

3D Vision Tasks

3D object detection
Bounding boxes for LiDAR / point clouds
Why AI teams choose CVAT for labeling their data
From Fortune 500 enterprises to world-class research labs to fast-moving AI startups, we help teams build great AI models, faster.

All annotation tools
under one roof

Annotate images, videos, and 3D models with a complete labeling toolkit. Use model-assisted labeling to quickly turn raw data into train-ready datasets.

Flexible deployment and infrastructure control

Run CVAT online, in your VPC, or fully air-gapped. Start from open source, scale with GPUs or Kubernetes, and integrate via APIs with support from our team.

Seamless integration into your ML stack

Connect internal tools, custom models, or external platforms. Use webhooks, batch auto-annotation, and model-assisted tools to accelerate labeling with automation.

Advanced team and project workflows

Structure labeling work from projects to tasks to jobs, and assign each to specific owners, reviewers, and annotators to streamline collaboration and keep track of the entire pipeline.

Quality assurance and analytics built in

Ship reliable datasets with validation via ground truth, honeypots, and cross-checks. Track throughput and quality with dashboards that scale from small teams to hundreds of users.

Enterprise-grade security and access control

Control who sees and does what with SSO (OIDC/SAML), granular RBAC, and audit logs. Enforce least-privilege access and meet internal security requirements.
Industries and applications
From 3D object detection in LiDAR to tumor segmentation in medical scans to product tagging in ecommerce—we’ve labeled it all.

Autonomous
Vehicles

Object detection, 3D LiDAR data labeling, lane marking, sensor fusion, pedestrian tracking.

Healthcare &
Medical Imaging

Anatomical structure segmentation, tumor detection, facial landmark annotation for diagnostics and research.

Retail &
Ecommerce

Product classification, shelf monitoring, logo detection, attribute tagging for catalogs.

Military &
Defense

Satellite and drone imagery annotation for object detection, activity monitoring, and infrastructure analysis in conflict zones.

Manufacturing
& Robotics

Defect detection, part recognition, pose estimation, object tracking in assembly lines.

Surveillance
& Security

Action recognition, people tracking, event tagging, crowd behavior analysis.

Agriculture

Plant disease detection, fruit counting, segmentation of crops from drone or field imagery.

Aerial & Satellite
Imagery

Building detection, land use classification, and road and infrastructure annotation.

Research &
Academia

Custom annotation projects for novel computer vision tasks in leading labs and institutions.
Labeling Services
Professional data labeling services for AI and machine learning teams.

Scalable Workforce with Quick Turnaround

Our in-house labeling team consists of 400+ experienced annotators across 12 time zones, ready to handle projects of any size on your schedule.

Quality-Assured, Secure Delivery

Each project is managed end-to-end with a dedicated project manager and measurable quality KPIs (accuracy, consistency) to ensure annotation excellence.

Fast Project Kick-Off, On-Time Delivery

We start within 1–2 days after contract signing and deliver high-quality labeled data on your timeline—even for large, complex projects.
Contact us

Get in Touch
with

Our Team