Hindi Image Paragraph Captioning Dataset (Translated and Corrected from Stanford Image-Paragraph Captioning Dataset)

This dataset is a Hindi-translated and corrected version of the Stanford Image-Paragraph Captioning Dataset, designed to support research and development in multilingual and low-resource image captioning tasks. It includes detailed paragraph-level captions for images, which describe various visual elements and interactions in natural scenes. The original dataset was developed by Ranjay Krishna et al. at Stanford University (Original source) and is available via Kaggle (Kaggle dataset link).