In April 2023, we introduced the original SAM model to our platform a mere few days after its initial release. Today, we are excited to announce the integration of Meta’s new Segment Anything Model, SAM 2, into our automated labelling suite, just one day after its official release. This rapid integration underscores our commitment to providing our customers with access to cutting-edge machine learning techniques faster than ever before.

Integrating SAM 2 brings enhanced accuracy and speed to your automated segmentation workflows, enhancing both throughput and user experience. We’re starting today by bringing SAM 2 into image segmentation tasks, where it’s been benchmarked to perform up to 6x faster than SAM. We are also looking forward to introducing the VOS capabilities of SAM 2, enhancing performance on automating video segmentation technologies already in Encord, such as SAM + Cutie.

Supercharge Your Annotations with the

Segment Anything Model

Label with SAM in Encord

As an extremely new piece of technology, SAM 2 is being made available to all our customers via Encord Labs. To enable SAM 2, navigate to Encord Labs in your settings and enable the switch for SAM 2, as illustrated in our documentation. When you return to the editor, you’ll know SAM 2 is enabled by the enhanced magic wand icon in the editor, signalling that you are using the latest and most powerful tools for your annotation tasks.

We are eager for our customers to try out SAM 2 and experience its benefits firsthand. We believe that this integration will significantly enhance the capabilities of our platform and provide unparalleled accuracy and speed in data annotation. We invite all users to send their feedback to product@encord.com. Your insights are invaluable as we continue to push the boundaries of what’s possible in machine learning annotation and evaluation.

Thank you for being a part of this exciting journey with Encord. We look forward to continuing to deliver world-leading technology at a rapid pace to meet the needs of our innovative customers.