跳到主要内容

翻译状态： 此任务示例目前仅提供英文版本。

Tagging Annotation Example

Tagging assigns descriptive labels to an entire image or to selected shapes.

Tagging Model​

X-AnyLabeling includes RAM and RAM++ for image- or shape-level tagging. Tagging models can also be combined with detection and segmentation models, as in GroundingSAM.

Here:

  • RAM is a strong image tagging model, which can recognize any common category with high accuracy.
  • RAM++ is the next generation of RAM, which can recognize any category with high accuracy, including both predefined common categories and diverse open-set categories.

Image-level Tagging​

RAM and RAM++ write their ordered predictions to the top-level tags array.

The Image Tags panel opens automatically after a tagging result and supports adding, editing, deleting, and drag-and-drop reordering.

"tags": ["person", "car", "street"]

Shape-level Tagging​

For detailed output examples, including the structured image-level tags field, refer to this folder.