This dataset contains real-world handwritten examination answer scripts collected from 21 students, scanned using mobile phone cameras. The images were manually organized into 21 class folders, corresponding to individual students.
The dataset includes 5,914 images, all resized to 224×224 pixels, and contains natural variations such as rotation and flipping, making it suitable for training robust… See the full description on the dataset page:
https://huggingface.co/datasets/anshcode1/AnswerScripts.
