GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → Kyung-Min → PororoQA

Kyung-Min / PororoQA

Licence: other
PororoQA, https://arxiv.org/abs/1707.00836

Labels

datasetvideo-processingquestion-answeringvideo-story-qa

Projects that are alternatives of or similar to PororoQA

Deep Embedded Memory Networks
https://arxiv.org/abs/1707.00836
Stars: ✭ 19 (-26.92%)
Mutual labels:  video-processing, question-answering
Acid.Cam.v2.OSX
Acid Cam v2 for macOS distorts video to create art.
Stars: ✭ 91 (+250%)
Mutual labels:  video-processing
CVSandbox
Open source software package, which aims to allow solving different tasks related to computer vision areas, like video surveillance, vision based automation/robotics, different sorts of image/video processing, etc.
Stars: ✭ 56 (+115.38%)
Mutual labels:  video-processing
iPerceive
Applying Common-Sense Reasoning to Multi-Modal Dense Video Captioning and Video Question Answering | Python3 | PyTorch | CNNs | Causality | Reasoning | LSTMs | Transformers | Multi-Head Self Attention | Published in IEEE Winter Conference on Applications of Computer Vision (WACV) 2021
Stars: ✭ 52 (+100%)
Mutual labels:  question-answering
NCE-CNN-Torch
Noise-Contrastive Estimation for Question Answering with Convolutional Neural Networks (Rao et al. CIKM 2016)
Stars: ✭ 54 (+107.69%)
Mutual labels:  question-answering
FFCreatorLite
一个基于node.js的轻量极速短视频加工库 A lightweight and fast short video processing library based on node.js
Stars: ✭ 155 (+496.15%)
Mutual labels:  video-processing
video-colorization
Applying the machine learning tool "DeOldify" on videos, frame by frame.
Stars: ✭ 15 (-42.31%)
Mutual labels:  video-processing
finance-qa-spider
金融问答平台文本数据采集/爬取，数据源涉及上交所，深交所，全景网及新浪股吧
Stars: ✭ 33 (+26.92%)
Mutual labels:  question-answering
denspi
Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index (DenSPI)
Stars: ✭ 188 (+623.08%)
Mutual labels:  question-answering
patrick-wechat
⭐️🐟 questionnaire wechat app built with taro, taro-ui and heart. 微信问卷小程序
Stars: ✭ 74 (+184.62%)
Mutual labels:  question-answering
MLH-Quizzet
This is a smart Quiz Generator that generates a dynamic quiz from any uploaded text/PDF document using NLP. This can be used for self-analysis, question paper generation, and evaluation, thus reducing human effort.
Stars: ✭ 23 (-11.54%)
Mutual labels:  question-answering
ODSQA
ODSQA: OPEN-DOMAIN SPOKEN QUESTION ANSWERING DATASET
Stars: ✭ 43 (+65.38%)
Mutual labels:  question-answering
QA4IE
Original implementation of QA4IE
Stars: ✭ 24 (-7.69%)
Mutual labels:  question-answering
avconv4java
Java interface to avconv tool
Stars: ✭ 26 (+0%)
Mutual labels:  video-processing
Image Processing
Image Processing techniques using OpenCV and Python.
Stars: ✭ 112 (+330.77%)
Mutual labels:  video-processing
Smart-Traffic-Signals-in-India-using-Deep-Reinforcement-Learning-and-Advanced-Computer-Vision
We have used Deep Reinforcement Learning and Advanced Computer Vision techniques to for the creation of Smart Traffic Signals for Indian Roads. We have created the scripts for using SUMO as our environment for deploying all our RL models.
Stars: ✭ 131 (+403.85%)
Mutual labels:  video-processing
vidpipe
Video data processing pipeline using OpenCV
Stars: ✭ 33 (+26.92%)
Mutual labels:  video-processing
GAR
Code and resources for papers "Generation-Augmented Retrieval for Open-Domain Question Answering" and "Reader-Guided Passage Reranking for Open-Domain Question Answering", ACL 2021
Stars: ✭ 38 (+46.15%)
Mutual labels:  question-answering
deformer
[ACL 2020] DeFormer: Decomposing Pre-trained Transformers for Faster Question Answering
Stars: ✭ 111 (+326.92%)
Mutual labels:  question-answering
PersianQA
Persian (Farsi) Question Answering Dataset (+ Models)
Stars: ✭ 114 (+338.46%)
Mutual labels:  question-answering
View All Similar Projects ➔

PororoQA

Authors: Kyung-Min Kim, Min-Oh Heo, Seong-Ho Choi, and Byoung-Tak Zhang (Seoul National University & Surromind Robotics)
Paper1: DeepStory: Video Story QA by Deep Embedded Memory Networks (https://arxiv.org/abs/1707.00836) (IJCAI 2017)
Paper2: PororoQA: A Cartoon Video Series Dataset for Story Understanding (NIPS 2016 Workshop on Large Scale Computer Vision System, 2016)

Download Video Clips

You can download the whole Pororo video clips with this URL.
If this link doesn't work, please email me [email protected]

Why cartoon video seires?

Characteristic of “Pororo”, children's cartoon video series

  • Easy sentence
  • A simple story structure
    • One episode: Avg. 7.2 min.
    • Similar events are repeated
  • Small environments
    • The size of dictionary: 3,729
    • The number of characters: 13
  • It may be appropriate as a test bed for video story QA problem

Data Statistics - QAs

  • 8,834 QA pairs
  • Remove 'ambiguous', 'unrelated' questions that do not follow guidelines
  • Average QA count per episode : 51.66
  • Examples and statistics by type of question Alt text
  • Statistics about questions based on answer types Alt text
  • Average number of words in our dataset depend on the first word in the question Alt text
  • Comparison of various public datasets in terms of video story analysis Alt text

Data Statistics – Scene Descriptions

  • 27,328 scene descriptions
    • Average sentence / word count per scene description : 1.7 / 13.6
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].