Automated labeling

Encord provides several distinct automated labeling techniques to help you create labels fast and with little effort:

  1. Segment anything model (SAM) allows you to automatically create labels around distinct features in all supported file formats.

  2. Interpolation is an automated process to estimate the location of labels between two manually created labels. Simply put, it is "filling in the blanks" between labels you've created.

  3. Object tracking is a process to estimate the location of a label based on pixel information enclosed within the labels you have manually provided.

  4. Auto segmentation tracking accurately tracks polygon and bitmask labels across a range of frames.



If you're looking to build and train models to detect specific objects, check out our Encord Apollo documentation!

What's the difference between models and automated labeling techniques?

There are key differences to understand between using models or the automated labeling techniques outlined above.

  • Models need to be trained to 'learn' how to identify specific objects or classifications. Once a model has been trained, it is able to automatically label a dataset without needing manual labels as an input. One model can be used across many projects.

  • Automated labeling, as outlined above, is much simpler and uses information provided by manual labels to automatically estimate the location of missing labels.