Abstract: This paper introduces a groundbreaking enhancement to image captioning through a unique approach that harnesses the combined power of the Vision Encoder-Decoder model. By leveraging the Swin ...