چکیده مقاله
Image captioning is the process of creating a description for an image Recognizing the significant items, their characteristics, and their relationships in a picture is necessary for image captioning Additionally, it must produce phrases that are both syntactically and semantically sound We give a thorough analysis of image captioning, encompassing both classical approaches and more recent deep learning techniques, to outline the most recent developments in image captioning The early classical works based on retrieval and template are specifically reviewed in the first brief section Then the research is concentrated on deep learning image captioning, which is divided into the encoder decoder architecture, attention mechanism, dense captioning, and semantics strategies We also discuss the datasets and evaluation metrics that are frequently used in automatic image captioning that are based on classical machine learning and deep learning methods
کلیدواژهها
نویسندگان
Mehrnaz Yousefi
Software Process Manager of Neshan Map and Navigator
شیوه ارجاع
Yousefi, Mehrnaz,1401,A survey on image captioning: An overview of methods, datasets and evaluation metrics,The 15th National Conference of Electrical, Computer and Mechanical Engineering,Shirvan
ارائهشده در
مجموعه مقالات شانزدهمین کنفرانس ملی مهندسی برق، کامپیوتر و مکانیک29 آذر 1401 · شیروان