Yes, we provide a fully managed data labeling service for Team and Enterprise plan users. We have partnerships with a few verified labeling agencies that can guarantee highest quality labeling workforce. and we distribute your labeling tasks according to your specifications. Contact us to start a fully managed labeling project!

Use to open source your datasets for the ML community around the world. We believe datasets, like ML models, should be more actively open sourced in order to lower the entry barrier for teams looking to build AI/ML. The storage size for the Free plan should be plenty to host various datasets. Private projects are for hosting internal, private datasets that only invited users can access.

Many proprietary, patent-pending deep learning technologies are used in Auto-Labeling. For example, Bayesian Deep Learning and Uncertainty Estimation techniques are used to measure labeling confidence levels, and class-agnostic refinement techniques improve auto-labeling accuracy. We can auto-label bounding box and polygon segmentations for around a hundred common object classes. Read more about our tech on our blog, and stay tuned for more exciting updates!

If you are an Enterprise plan customer, we may be able to support custom annotation types for you, as long as it is something on our product roadmap. Please contact us for any inquiries on supporting custom annotation or data types.

We do have plans to support on-premise solutions on private clouds for Enterprise plan customers later this year. Contact us and let us know your needs, and we’ll reach out to you when everything is ready.

We support image data for all usage plans, and support video data for enterprise customers. If you are a Team Plan user and would like access to our Video support, please contact us for more information. Support for other data types, including text and Lidar is coming soon later this year.