Hackathon Posts - Oracle AI Service: Vision

 

 Few posts planned for use in expected Hackathons. 

We want to use Oracle AI  Service of Vision. Please note there is a specific Document Understanding Service as well.

OCI Vision is an AI service for performing deep-learning–based image analysis at scale. With prebuilt models available out of the box, developers can easily build image recognition and text recognition into their applications without machine learning (ML) expertise. For industry-specific use cases, developers can automatically train custom vision models with their own data. These models can be used to detect visual anomalies in manufacturing, organize digital media assets, and tag items in images to count products or shipments.

This is the documentation home, with most of the essentials.

Permissions

First, we need to gain access to the service. It can be granted to all OCI users of the tenancy or specific group you belong to. For this we need to create a Policy that allows to use ai-service-vision-family in tenancy.

    To create a Policy, in the burger menu go to "Identity & Security" and select Policies. There press "Create Policy". Once you did enter Name and description of the Policy and turn on the "show manual editor" under Policy Builder. Now you have 2 options to allow use for specific group or all the users. Enter one of the following:

•  allow group <group-name> to use ai-service-vision-family in tenancy

• allow any-user to use ai-service-vision-family in tenancy 
  

 

For vision we probably need access to object storage where the source files can be located.

In Hackaton we might be careless and grant the general access to object storage:

• allow group <group_name> to use object-family in tenancy

 In real life we will probably restrict it to a specific compartment:

• allow group <group_name> to use object-family in compartment <input_bucket_located_object_storage_compartment>

And a policy to access output location in object storage of a specific compartment

• allow group <group_name> to manage object-family in compartment <output_bucket_located_object_storage_compartment>

 

In case you plan to use later the Oracle Cloud shell, you might want to do the same with "cloud-shell in tenancy". For example: allow any-user to use cloud-shell in tenancy 

 

 Intro

Now we can access the AI Service - Vision (under the Analytics & AI menu).

 

I really liked the way the actual product in OCI seems to be self documented:

 

Links

 

• The service Playlist here Including 47 minutes intro, Guide how to set up policies and a creating a custom model. 

• A link to Oracle LiveLabs workshop here.
• A liveLabs workshop on "Use Data Labeling Service to Create a Biomedical Image Classification Model" 
  
• Policies documentation that includes more information that described above.
  

• Service Overview document
• Vision API Documentation (details in the menu on the left)
  

• Building a custom Model in Console 
  
• Pretrained AI Models Documentation

• Analyse video with Python Sample code
• Integrating Vision with Oracle Analytics that should point here
  

 

At the bottom of the page we see the Service capabilities:

• Image classification - Categorizes scene-based features and objects within an image
• Object detection - Locates and identifies objects within an image
• Text detection - Provides word-level and line-level text as well as the bounding box coordinates of where the text is located
• Custom image classification and object detection - Image classification and object detection models trained on your own labeled dataset
  
  
  
  

On the left part of the screen they are the link to use the console UI:

 

 See also:

• Announcement Blog here
• Product website here