Skip to content
Nappai Documentation

Describe Image

This component uses artificial intelligence to describe images in your Nappai automation system. You can use it to automatically generate text that explains what’s in an image. This is helpful for tasks like creating image captions, summarizing visual data, or adding context to images in your reports.

Relationship with AI Language Models

This component works by using a powerful AI language model to analyze images and generate descriptions. The quality of the description depends on the chosen language model and the prompt you provide.

Inputs

  • Prompt Text: This is where you provide instructions to the AI on how to describe the image. For example, you could type “Describe the image in detail,” or “Summarize the main objects in the image.” A default prompt is provided, but you can customize it for more specific results.
  • Image Data: This is where you provide the image you want described. You can upload the image directly or provide a link to it.
  • Language Model: You need to select the AI language model you want to use for image analysis. Nappai will provide a list of available models. The choice of model can affect the quality and style of the description.
  • Resize Image: This option lets you choose whether to resize the image before it’s processed. Keeping the original size is usually best, but you can choose smaller sizes (255x255 or 512x512 pixels) if you need to speed up processing or if your image is very large.

Outputs

The component produces a single output:

  • Data: This output contains the AI-generated description of your image. This text can then be used in other parts of your Nappai workflow, such as adding it to a document, sending it in an email, or displaying it on a dashboard.

Usage Example

Imagine you’re building a system to automatically caption images for your social media posts. You could use the “Describe Image” component to analyze an uploaded image, and then use another Nappai component (like “LinkedIn Post” or “Slack Message”) to automatically post the image along with the generated caption.

Templates

This component is used in the following Nappai templates:

  • AI-Powered Property Description Optimizer
  • TicketToJSON

Several Nappai components work well with “Describe Image”:

  • Summarizer: Use this to shorten the generated description if needed.
  • LinkedIn Post: Post the image and its description directly to LinkedIn.
  • Google Sheet Writer: Save the image description to a Google Sheet for later analysis.
  • Many other components: The “Data” output can be used as input for a wide variety of Nappai components, allowing you to integrate image description into many different workflows. (See the full list of connected components in the technical documentation).

Tips and Best Practices

  • Experiment with prompts: Try different prompt texts to get the best results. More specific prompts usually lead to more accurate descriptions.
  • Choose the right language model: Different language models have different strengths. Experiment to find the model that best suits your needs.
  • Consider image resizing: If you have very large images, resizing them might speed up processing without significantly impacting the quality of the description.

Security Considerations

Ensure that the images you upload do not contain sensitive or confidential information. The AI model processes the image data, so be mindful of data privacy. Refer to Nappai’s overall security documentation for more details.