Similar Items: Contextual image caption creation using object positional embedding and generative models