Overview

Object detection models return bounding boxes. These boxes indicate where an object of interest is in an image. In many models, such as Ultralytics YOLOv8, bounding box coordinates are horizontally-aligned. This means that there will be spaces around angled objects.

‍

Example of Orient Bounding Boxes (Image 2 uses OBB). Sourced from Github Thread.

‍

You can retrieve bounding boxes whose edges match an angled object by training an oriented bounding boxes object detection model, such as YOLOv8's Oriented Bounding Boxes model.

‍

Use Cases

‍

Some objects need to be detected in certain ways. In the image above, while the first bounding boxes are accurate, they may not be helpful towards the user due to the messiness of the outputted image. However, in the second image, through the use of bounding boxes the resistors are detected in a neat and orderly fashion, making the data a lot easier to work with.

‍

In the real world, objects will not always be straight, so the application of oriented bounding boxes can help create neater computer vision models for many more use cases.

Oriented Bounding Boxes in a Real-World Situation. Sourced from this Paper.

Performance

Use This Model

Label Data Automatically with YOLOv8 Oriented Bounding Boxes

You can automatically label a dataset using YOLOv8 Oriented Bounding Boxes with help from Autodistill, an open source package for training computer vision models. You can label a folder of images automatically with only a few lines of code. Below, see our tutorials that demonstrate how to use YOLOv8 Oriented Bounding Boxes to train a computer vision model.

No items found.