Intersection over Union (IoU)

Encord Computer Vision Glossary

Intersection over Union (IoU) is a commonly used performance metric in object detection tasks in machine learning. It measures the overlap between the predicted bounding box of an object and the ground truth bounding box in an image.

In object detection, a bounding box is used to localize and identify objects within an image or video frame. The IoU metric compares the area of intersection between the predicted bounding box and the ground truth bounding box with the area of union between the two boxes. It is calculated as the ratio of the intersection over the union, and its value ranges from 0 to 1, where 1 indicates a perfect match between the predicted and ground truth bounding boxes.

IoU is a crucial metric for evaluating the accuracy of object detection algorithms in machine learning. It measures how well the predicted bounding box aligns with the ground truth and how precise the object detection algorithm is in localizing objects. A high IoU score indicates a good detection performance, while a low score indicates poor detection performance.

background image

One platform for creating better training data and debugging models.

IoU is used to evaluate the performance of various object detection algorithms, including Faster R-CNN, YOLO, and SSD. It is also used in various computer vision tasks such as pedestrian detection, vehicle detection, and face recognition. In addition, IoU is often used in conjunction with other performance metrics such as precision, recall, and F1 score to provide a comprehensive evaluation of the object detection algorithm.

In summary, 

  • Intersection over Union (IoU) is a critical performance metric used in object detection tasks in machine learning. 
  • It measures the accuracy of the predicted bounding box compared to the ground truth bounding box and is widely used in various computer vision applications.
  • A high IoU score indicates a good detection performance, while a low score indicates poor detection performance.
cta banner

Get our newsletter to learn about the latest developments in computer vision