Abstract
Recurrent neural networks have recently emerged as a useful tool in computer vision and language modeling tasks such as image and video captioning. The main limitation of these networks is preserving the gradient flow as the network gets deeper. We propose a video captioning approach that utilizes residual connections to overcome this limitation and maintain the gradient flow by carrying the information through layers from bottom to top with additive features. The experimental evaluations on the MSVD dataset indicate that the proposed approach achieves accurate caption generation compared to the state-of-the-art results. In addition, the proposed approach is integrated with our custom-designed Android application, WeCapV2, capable of generating captions without an internet connection.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
