This repository contains the main Google Colab file used for this research. The dataset used for this project is EDUVSUM, which can be accessed here https://zenodo.org/records/4002959#.X0bsiBmxXmF.
Additionally, I have attached a file called Ground_truth_query_vs_vtt.xlsx. This Excel sheet contains a benchmark dataset created by the model expert team and has been made public to assist new researchers in validating their video summarization models.
Feel free to customize further if needed! [email protected]