This dataset is made up of research images from different domains, annotated according to their content.

Each main folder is associated with a researcher, where the number that identifies it is labeled (R1 = Researcher 1).

Each of the main folders is divided into new folders that contain images and their respective descriptions. (Ex: R2_I1 = Researcher 2 Image 1).

In the text document, the description elements are divided into parts:

Formal descriptors - the folder designation corresponding to the image-description pair, the file name, the mention of authorship of the image, and the researcher who provided and described the image.

Content descriptors - obtained through image content analysis. 

Formal descriptors are separated by semicolons. Content descriptors are separated by a comma. Between formal descriptors and content descriptors a semicolon is used.

Example:

R6_I1; Enriched cage.jpg; Anna Olsson; Anna Olsson; Mus musculus, mouse cage, environmental enrichment, mice, laboratory animals