Overview
This article discusses the +AI Vision feature, which enables users to upload an image and receive answers to specific inquiries regarding the contents of the image. This functionality allows users to obtain pertinent information about the image without the necessity of opening and examining it personally. In this article, you will learn how to configure and utilize +AI Vision.
Note: The +AI Vision feature has a maximum image file size limit of 20MB. Only one image may be uploaded when +AI Vision is enabled. Supported image formats include PNG, JPEG, JPG, WEBP, and GIF. An OpenAI license is required to utilize AI features on SmartSimple. Additionally, only a Global Administrator can enable this feature. For further information regarding billing and implementation, please contact your account manager or sales@smartsimple.com.
For a general overview of the +AI Vision feature, please view this video.
To view this video in full screen, please click on the full screen button located at the bottom right.
Configuration
To configure +AI Vision, it is first necessary to create a custom field that will store the responses to the questions posed to the AI.
- Create a new custom field with the Field Type set to “Text Box - Multiple Lines.” Under Caption, enter “AI Vision Text.”
- (Optional) Enable the Label as AI-Generated Content option.
- Click Save.
Next, we will create an upload custom field that will utilize the +AI Vision feature:
- Create a new custom field with the Field Type set to “Upload - Multiple Files Storage.”
- In the File section, enable the Enable +AI Vision option.
- Input a relevant prompt into the +AI Vision Prompt field to inquire of +AI Vision regarding the upload. For instance, if you would like +AI Vision to provide a brief description of the contents of the uploaded image, enter “Describe what the image contains in 100 words or less.”
- Select the “Text Box - Multiple Lines” custom field created in Step 1 from the To Custom Field dropdown located in the Value Storage section. This instructs +AI Vision to store the responses to the +AI Vision Prompt in this custom field.
Use Case: Generate Descriptions for User-Uploaded Images
As an illustration, we will utilize +AI Vision to assist in automating a standard review process that involves user-generated image submissions. By employing +AI Vision, reviewers can access text summaries regarding the image submissions without the need to download or open the image files.
To activate +AI Vision and to test our previous configuration, please follow these steps:
- Create or edit a new record.
- Upload an image to the upload field with Enable +AI Vision activated. Wait for the image upload to complete.
- Click Save.
The text box field that we created to store responses to +AI Vision prompts will be populated with an answer.
This indicates that when a single image is uploaded to any upload field with +AI Vision enabled, the AI will automatically generate a response to our prompts. Here are some sample prompts to assist you in getting started:
- “Does the image contain at least one person?”
- “Describe the main color composition of the image.”
- “How effectively does the image utilize negative space?”