FirstSpirit AI Suite
Documentation

Image Assistant

The Image Assistant is an AI-based tool for image processing tasks in the ContentCreator. It automates two tasks that would otherwise require manual intervention for every uploaded image:

The assistant comes with a preconfigured set of prompts that allow you to get started right away. To harness the full potential of the assistant, customize existing prompts or create your own.

The configuration of FirstSpirit AI is mandatory before any assistant can be used. For more information see Connect a model and vendor.
The corresponding module has to be installed and the relevant project component configured before the assistant can be used. For more information see Configuring the AI Suite.

Automatic Image Cropping

When an image is uploaded to the Media Management area, the module analyzes its content using AI-assisted image recognition. Detected subjects are prioritized in the following order by default:

  • A single face, if recognized, is used as the focus area.
  • Multiple faces are framed together as the focus area.
  • A single object is used as the focus area.
  • Multiple objects are framed together as the focus area.

Based on the determined focus area, FirstSpirit automatically crops the image for all configured resolutions. Editors can review and adjust automatic crops manually at any time.

To manually trigger an analysis and cropping for an already uploaded image:

  • Navigate to Media Management area.
  • Hover over the image you want to trigger a re-analysis for.
  • Click the Rescan-button .

For more information in Automatic Image Cropping see here.

Generating alt-text

When an image is uploaded and alt text generation is configured, alt text is generated automatically using the connected model and the configured prompt.

Alt text generation produces output in a single language, because image metadata holds one value per field across all project languages. The output language is determined by the configured prompt. To change the output language, adjust the prompt accordingly.

To access the generated text:

  • Navigate to the Media Management area.
  • Hover over the image you want inspect the metadata for.
  • Click the Set metadata-button .
    → A new window opens.
  • Edit the metadata of the image in the corresponding text field.

To manually trigger the generation of alt text for an image or folder:

  • Navigate to Media Management area.
  • Hover over the image or folder you want trigger alt text generation for.
  • Click the Rescan-button .

Rescanning an image overwrites any previously entered alt text with the newly generated one. Rescanning a folder generates alt text for all images in the folder that do not yet have an alt text.

Limits

  • Image generation is not supported. The Image Assistant analyzes existing images only.
  • Only JPEG, PNG, and WebP images are analyzed. Images in other formats are not processed, and automatic cropping and focus area detection are not available for them.
  • Alt text output is single-language. The output language is determined by the configured prompt.

  • Image clarity and resolution affect the quality of the analysis result. Blurry or low-resolution images may produce incomplete or inaccurate results.

Table of Contents