Authors :
Hansaraj Wankhede; R Bharathi Kumar; Sushant Kawade; Ashish Ramtekkar; Rachana Chawke
Volume/Issue :
Volume 8 - 2023, Issue 4 - April
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://bit.ly/3VtJFVS
DOI :
https://doi.org/10.5281/zenodo.7888972
Abstract :
- To accomplish video summarization, one must
possess fundamental comprehension and assessment
skills regarding the content. In this project, an AI-based
video summarization system using FFmpeg, Natural
Language Processing (NLP) techniques, and AssemblyAI
has been developed. The system aids in generating an
accurate summary. We analyse previous works and
suggest a fair data split for future reference. The parallel
attention mechanism that utilizes static and motion
features significantly improves the results for the
SumMe dataset and performs well for other datasets as
well. The primary aim is to provide users with an
efficient way to comprehend video content, and the
system's effectiveness is evaluated based on the accuracy
and comprehensiveness of the generated summaries and
user satisfaction with the system's functionality.
Keywords :
AI-based, Video Summarization, FFmpeg, NLP, AssemblyAI, Static Features, Motion Features, SumMe Dataset, Accuracy, Comprehensiveness, user Satisfaction, Data Split, Benchmarking.
- To accomplish video summarization, one must
possess fundamental comprehension and assessment
skills regarding the content. In this project, an AI-based
video summarization system using FFmpeg, Natural
Language Processing (NLP) techniques, and AssemblyAI
has been developed. The system aids in generating an
accurate summary. We analyse previous works and
suggest a fair data split for future reference. The parallel
attention mechanism that utilizes static and motion
features significantly improves the results for the
SumMe dataset and performs well for other datasets as
well. The primary aim is to provide users with an
efficient way to comprehend video content, and the
system's effectiveness is evaluated based on the accuracy
and comprehensiveness of the generated summaries and
user satisfaction with the system's functionality.
Keywords :
AI-based, Video Summarization, FFmpeg, NLP, AssemblyAI, Static Features, Motion Features, SumMe Dataset, Accuracy, Comprehensiveness, user Satisfaction, Data Split, Benchmarking.
