Custom Vision Service

The video tutorial discusses the differences between Microsoft's Computer Vision and Custom Vision services. While Computer Vision offers pre-built models for general use, Custom Vision allows users to create domain-specific models by uploading and tagging images. The tutorial explains the process of building and training custom models, highlighting the two main features of Custom Vision: classification and object detection. It also provides practical examples and addresses common questions related to the service.

What is the main advantage of using Custom Vision Service over Computer Vision Service?

• It allows for the creation of domain-specific models.

• It is cheaper to use.

• It requires no training data.

• It has more pre-built features.

Which step is NOT involved in training a custom model using Custom Vision Service?

• Writing code for model training

• Using tagged images for training

• Tagging images

• Uploading images

What are the two main features provided by the Custom Vision Service?

• Facial recognition and segmentation

• Classification and object detection

• Clustering and segmentation

• Thumbnail generation and color scheme detection

If you need to build a butterfly image classifier using your own images, which Azure service should you use?

• Text Analytics

• Face API

• Custom Vision Service

• Computer Vision Service

Which project types are supported by the Custom Vision Service?

• Facial recognition and clustering

• Classification and object detection

• Segmentation and clustering

• Thumbnail generation and color scheme detection