AVQANet: Amharic Visual Question Answering Dataset for Ethiopian Museum Artifacts

This dataset was developed for research on Amharic Visual Question Answering (AVQA) systems for Ethiopian cultural heritage artifacts. The dataset contains artifact images and corresponding Amharic question-answer pairs collected from Ethiopian museums to support multimodal deep learning research in low-resource languages particularly in museum domain. The dataset was created to facilitate the development of intelligent museum guidance systems, cultural heritage preservation technologies, and Amharic-language multimodal artificial intelligence applications. The dataset was collected from the National Museum of Ethiopia (NME) and Lake Tana monasteries museum by capturing slit lamp camera. The dataset includes: Artifact images Amharic questions Ground-truth answers Training and testing splits Annotation files This dataset supports the research presented in the paper entitled: “AVQANet: Amharic Visual Question and Answering Model Based on Deep Learning Approach for Ethiopian Museum Visitors”. The dataset includes artifact images and Amharic question-answer annotations used for training and evaluating the proposed AVQANet framework.