EVOLUTION OF AUTOMATIC VISUAL DESCRIPTION TECHNIQUES - A SURVEY

Abstract

Describing the contents and activities in an image or video in semantically and syntactically correct sentences is known as captioning. Automated captioning has been one of the most competitive major trends in present day research with new sophisticated models being discovered every day. Captioning models require intense training and perform intense complex calculations before successfully generating a caption and hence, takes considerable amount of time even in machines with high specifications. In this survey, we go through the recent state-of-the-art advancements in automatic image and video description methodologies using deep neural networks and summarize the important concepts that can be inferred from the researches. The summarization has been done with detailed analysis of methodologies used along with explanation of referenced context. Along with detailed description of available datasets and methodologies. The focus of our research lies in techniques which are able to optimize existing concepts as well as incorporate new methods of visual attention to generate captions. This survey emphasizes on the importance of applicability and effectiveness of existing works in real life applications and highlights those computationally feasible and optimized techniques which can be supported in multiple devices ,including lightweight devices like smartphones.