Summarization is an ongoing field of study with the objective of providing a concise and useful gist from vast amounts of content. Everything is digitalized these days, and digital data is utilized for various purposes on the Internet. There are diverse forms of content, ranging from different formats to domains and topics. E-learners find it relatively tough to read and understand the lengthy content. To address the aforementioned challenge, this research study has proposed a novel summarization approach based on different characteristics and features of summaries that can flexibly switch between extractive, abstractive, and illustrative summaries with images and videos. The proposed Deep Wrap Up provides extractive summary by using the NLTK (Natural Language Toolkit) library, abstractive summary using T5 (Text-to-Text Transfer) transformers, summary with images using the CLIP (Contrastive Language-Image Pre-Training) library, and summary with videos generated using moviePy. As a result, our technique successfully combines the benefits of several branches of summarization, balancing informativeness and conciseness.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    DEEP WRAP-UP- Automatic Document Summarization with Animations




    Publication date :

    2022-12-01


    Size :

    1360922 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Animations

    Nathan Mahir Schilling | NTRS


    Jason-1 Orbit Animations

    Kunstman, G. | NTRS | 1998



    Systematic Study of Extractive Multi-Document Summarization Techniques

    Ulaganathan, Nagul / Rohith, J / Aravind, Sri et al. | IEEE | 2020


    Correlation Animations and Volume Detections of a Deep Marine Fan System

    Norwegian Petroleum Society (NPF) | British Library Conference Proceedings | 1999