Constructing Hierarchical Q&A Datasets for Video Story Understanding

Title
Constructing Hierarchical Q&A Datasets for Video Story Understanding
Author(s)
허유정; 온경운; 최성호; 임재서; 김진아; 류제광; 배병철; 장병탁
KIOST Author(s)
Kim, Jin Ah(김진아)
Publication Year
2019-03-15
Abstract
(arXiv.org: https://128.84.21.199/abs/1904.00623)

Video understanding is emerging as a new paradigm for studying human-like AI. Question-and-Answering (Q&A) is

used as a general benchmark to measure the level of intelligence for video understanding. While several previous

studies have suggested datasets for video Q&A tasks, they did not really incorporate story-level understanding, resulting in highly-biased and lack of variance in degree of question difficulty. In this paper, we propose a hierarchical method for buildingQ&A datasets, i.e. hierarchical difficulty levels.

We introduce three criteria for video story understanding, i.e. memory capacity, logical complexity, and DIKW (Data-Information-Knowledge-Wisdom) pyramid. We discuss how three-dimensionalmap constructed from these criteria can be used as a metric for evaluating the levels of intelligence relating to video story understanding.
URI
https://sciwatch.kiost.ac.kr/handle/2020.kiost/22805
Bibliographic Citation
AAAI 2019 Spring Symposium, pp.1 - 9, 2019
Publisher
AAAI (Association for the Advancement of Artifical Intelligence)
Type
Conference
Language
English
Publisher
AAAI (Association for the Advancement of Artifical Intelligence)
Related Researcher
Research Interests

AI/Machine Learning,Climate Change,Marine Disaster,인공지능/기계학습,기후변화,해양기상재해

Files in This Item:
There are no files associated with this item.

qrcode

Items in ScienceWatch@KIOST are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse