COMPARING ENCODER-DECODER NEURAL ARCHITECTURES FOR AUTOMATIC VIDEO CAPTIONING

There are a huge number of videos uploaded around the world from different areas at one second through Internet; these videos need to be arranged and classify based on their captions.

Considering the research done in the field of video captioning, the current research project tries to address a gap in the filed by introducing a solution to video captioning that is enhanced by DL models. This solution has practical applications as well. Primarily, and the demand for video understanding increased in domains such as surveillance, medicine, social, and military applications. Video captioning facilitates a simple summarizing, indexing, and searching data for various types of users including the visually impaired.