News
The bounding box placement is accurate down to a single pixel, allowing Figure Eight customers to have the most accurate possible human-driven object detection for their Computer Vision models.
Results that may be inaccessible to you are currently showing.
Hide inaccessible results