Towards fine-grained adaptive video captioning via Quality-Aware Recurrent Feedback Network

Tianyang Xu, Yunjie Zhang, Xiaoning Song, Xiao-Jun Wu

Abstract: Highlights•A Quality-Aware Recurrent Feedback Network is constructed to refine video features.•Predicted textual information is used in the process of progressively enhancing the video representation.•We verify the superiority of QARFNet in utilizing the generated captions via a feedback mechanism.