The Vision block is part of Airtable Blocks, a Pro plan feature. Blocks let you extend the functionality of your bases: you can use blocks to bring new information into Airtable, visualize and summarize your records in colorful ways, and even directly integrate your Airtable bases with your favorite apps.
With the Vision block, you can use the Google Cloud Vision API to get more insight out of images from your records. You can quickly classify your images into thousands of categories (like "dog," "lighthouse," or "Sahara"), extract those labels, and save them to a field in your base—meaning that you can tag hundreds of images with just a few clicks. You can also identify logos, or use optical character recognition (OCR) to extract text from images. Please note that this block requires a Google Cloud Vision API key.
Watch this video to learn more about how to set up and use a Vision block, or read on for further information.
After adding the block to your base, you'll be prompted to enter your Google Cloud API key.
Next, you'll need to select a table and view containing the records with the images you want to analyze, as well as select the specific attachment field containing the images.
Select at least one of the toggles representing the different ways that Google Cloud Vision can analyze your images.
Extract labels will output words or phrases based on how Google categorizes your images.
Extract logos will output the company or brand name associated with any logos depicted in your attachments.
Extract text will pull words from your images and put them in a designated text field.
For each Vision option you toggle, you'll need to select a destination text field into which the extracted text can go. You can select the destination text field from the dropdown menu that appears next to each toggled option.
When you're ready to analyze your images, click the Next button. This will bring up a confirmation dialog confirming your use of the API.
After you confirm by clicking the Analyze button, you'll see previews of the text that the block has extracted from your images on their associated records. Click the Save button to add the Vision block's output to the designated text field or fields.