What Is TrackForward?
TrackForward is a Plainsight AI-powered labeling feature that can dramatically reduce the time and effort required for labeling video data. Using AI, TrackForward analyzes the labels in one frame of a video to predictively label objects in subsequent frames. By labeling an object with either a Bounding Box or Polygon and selecting the TrackForward tool, Plainsight’s team can quickly generate labels automatically for desired objects across entire videos. TrackForward successfully labels images even as they move from frame-to-frame, making it useful for monitoring objects in motion, such as vehicles traveling through intersections, shoppers entering stores, and items in the production processes.
More Label Accelerators
TrackForward is just one of several Plainsight labeling features that accelerate dataset creation by 20X. These tools make data labeling faster and more efficient – often through the use of AI. Other accelerated labeling features that drive innovation for Plainsight customers include:
- CopyForward: Though it doesn’t involve AI, this feature reduces the tedium of repetitive labeling to a considerable degree. CopyForward automatically labels objects that remain stationary across multiple frames, eliminating the need to label the same object over and over again. It’s useful, for example, in instances where dwell time tracking is needed –how long a person or object remains in a specific location or within a video frame.
- SmartPoly: Bounding boxes are a useful type of label, but many times for instance and semantic segmentation or to define important image and object features, users need labels that adhere to the outlines of relevant objects. SmartPoly eliminates the need for repetitive clicking by automatically transforming bounding boxes into accurate polygon labels.
- AutoLabel: Using predictions from a machine learning model, AutoLabel can generate rectangle and polygon labels for and automatically apply them to images or videos. Labeling can be done using pre-trained models from the COCO dataset, a large dataset of common objects. Or, AutoLabel can be used with the Rectangle and Polygon label types and use a custom model to automatically customize object labeling.
Schedule a Call
Plainsight makes computer vision a catalyst for digital transformation for customers. Schedule a call to learn more about our proven approach.
