Video metadata generation is the process of extracting relevant information from a video and categorizing it into specific classes. The extracted data is used to help search engines and users in finding videos more easily, as well as to assist in organizing and archiving videos. The goal of this research is to provide an overview of the video metadata generation process. This paper outlines the key steps involved in video metadata generation and classification, including video analysis, text extraction, audio analysis, image analysis, classification, and metadata generation. Additionally, the report discusses the applications of video metadata generation and classification across various domains, such as entertainment, education, and business. The importance of video metadata generation and classification lies in its ability to make videos easily discoverable and accessible to users, leading to increased views and engagement. As video content continues to grow in popularity, video metadata generation and classification will become increasingly important for content creators and distributors to ensure their videos are easily discoverable and accessible to their target audience. Here, video metadata generation is done with the help of LCRN and ConvLSTM methods. And the performance in terms of accuracy is analyzed.