Datasets:
metadata
license: mit
task_categories:
- image-to-text
language:
- en
tags:
- Image-to-Text
- computer-vision
- captioning
pretty_name: Description Dataset
size_categories:
- n<1K
Description Dataset
Dataset Description
Languages
English (en)
long_description
: Detailed description of the imageshort_description
: Concise description of the imagelabel_cloth
: ID leather cloth imageDataset Structure
Each record in the dataset contains:
file_name
: Relative path to the image filetext
: Main description text used for traininglong_description
: Detailed description of the imageshort_description
: Concise description of the imagelabel_cloth
: ID leather cloth image
Created Files
Total Images
Dataset Creation
Source Data
The source images were processed using AI image captioning to generate natural language descriptions.