Computer vision in AI is a form of artificial intelligence that focuses on making computers interpret and understand the visual world. It enables them to recognize objects, detect movement, track objects, and respond to actions. There are several types of computer vision, which are described below:

1) Image Recognition: This type of computer vision involves identifying an object or scene in an image. The AI system interprets the image, detects features such as colors or shapes within it, and then assigns labels accordingly.

2) Object Detection: This type of computer vision is used to identify objects within a given image. It will draw bounding boxes around each detected object and label it according to its pre-trained categories.

3) Semantic Segmentation: This type of computer vision involves differentiating between pixels belonging to different objects present in an image. The goal is for the AI system to assign each pixel a class label based on what’s being seen in the picture (e.g., sky, dog).

4) Motion Detection: Motion detection can be used for surveillance purposes or other applications where tracking motion is important. The system will detect motion from one frame to another over time by analyzing changes between frames over time with respect to background subtraction methods and optical flow techniques etcetera..

5) Scene Understanding: Scene understanding allows machines to recognize entire scenes that consist of multiple objects under varying conditions like illumination or weather effects etcetera.. It involves locating all elements including people, buildings and vehicles while interpreting their environment accurately as well as understanding relations between them (like human-building interactions).