Using scene-specific contexts and region-based attention in neural image captioning
Primary LanguagePythonMIT LicenseMIT