Understand computer vision

Computer Vision is an area of AI that deals with visual processing. Let's explore some of the possibilities that computer vision brings.

The Seeing AI app is a great example of the power of computer vision. Designed for the blind and low vision community, the Seeing AI app harnesses the power of AI to open up the visual world and describe nearby people, text and objects.

View the following video to learn more about Seeing AI.

Computer Vision models and capabilities

Most computer vision solutions are based on machine learning models that can be applied to visual input from cameras, videos, or images. The following table describes common computer vision tasks.

Task                                     Description

Image Classification            

   

                                        Image classification involves training a machine learning model to classify images based on their contents

                                 . For example, in a traffic monitoring solution you might use an image classification model to

                                         classify images based on the type of vehicle they contain, such as taxis, buses, cyclists, and so on.

Object detection                 

                                       Object detection machine learning models are trained to classify individual                                           objects within an image, and identify their location with a bounding box. For                                             example, a traffic monitoring solution might use object detection to identify the location of different classes of vehicle.

Semantic segmentation       

                                          Semantic segmentation is an advanced machine learning technique in which                                            individual pixels in the image are classified according to the object to which                                            they belong. For example, a traffic monitoring solution might overlay traffic                                          images with "mask" layers to highlight different vehicles using specific colors.

Image analysis                 

                                    You can create solutions that combine machine learning models with advanced                                       image analysis techniques to extract information from images, including "tags"                                         that could help catalog the image or even descriptive captions that summarize the scene shown in the image.

Face detection, analysis,     
 

and recognition

                                        Face detection is a specialized form of object detection that locates human                                             faces in an image. This can be combined with classification and facial geometry                             analysis techniques to recognize individuals based on their facial features.

Optical character 

recognition (OCR)               

                                     Optical character recognition is a technique used to detect and read text in                                                 images. You can use OCR to read text in photographs (for example, road signs                                          or store fronts) or to extract information from scanned documents such as letters, invoices, or forms.

Computer vision services in Microsoft Azure

Microsoft Azure provides the following cognitive services to help you create computer vision solutions:

Service                  Capabilities

Computer Vision You can use this service to analyze images and video, and extract descriptions,                                         tags, objects, and text.

Custom Vision Use this service to train custom image classification and object detection models                                       using your own images

.

Face                 The Face service enables you to build face detection and facial recognition                                                 solutions.

Form Recognizer Use this service to extract information from scanned forms and invoices.

Next Page