Article Directory
Update time-2019.12 first draft
0. Introduction
The first step in learning VQA -pre-thesis research. Investigate the publication of papers at major conferences in recent years to understand the progress in this direction, including CVPR, ICCV, ECCV, ACM MM, and AAAI . After that, I am ready to summarize the commonly used data sets and classic methods.
1. ACM MM
ACM MM is a major international conference in the field of computer science and technology multimedia, focusing on the integration and processing of multi-angle information generated by different digital media. The VQA is part of its multimedia content understanding of the subject there (Understanding Multimedia Content ) The Vision and Language branch.
1.1 ACM MM 2019
- There are 5 incomplete statistics (including Video / Visual Question Answer)
Essay topic | Author |
---|---|
Multi-interaction Network with Object Relation for VideoQA | Zhejiang University |
Learnable Aggregating Net with Divergent Loss for VideoQA | University of Electronic Science and Technology |
Question-Aware Tube-Switch Network for VideoQA | University of Science and Technology of China |
CRA-Net: Composed Relation Attention Network for Visual QA | University of Electronic Science and Technology |
Erasing-based Attention Learning for Visual QA | Institute of Automation, Chinese Academy of Sciences |
1.2 ACM MM 2018
- There are 4 incomplete statistics (including Video / Visual Question Answer)
Essay topic | Author unit |
---|---|
Explore Multi-Step Reasoning in Video Question Answering | Tianjin University |
Fast Parameter Adaptation for Few-shot Image Captioning and Visual Question Answering | Southern University of Science and Technology |
Object-Difference Attention: A Simple Relational Attention for Visual Question Answering | Beijing University of Posts and Telecommunications |
Enhancing Visual Question Answering Using Dropout | Institute of Automation, Chinese Academy of Sciences |
1.3 ACM MM 2017
- There are 4 incomplete statistics (including Video / Visual Question Answer)
Essay topic | Author unit |
---|---|
VideoQA via Hierarchical Dual-Level Attention Network Learning | Zhejiang University |
VideoQA via Gradually Refined Attention over Appearance and Motion | Zhejiang University |
2. CVPR
CVPR stands for Conference on Computer Vision and Pattern Recognition, and the Chinese name is International Conference on Computer Vision and Pattern Recognition, which is usually held around June every year.
2.1 CVPR 2019
- There are 12 incomplete statistics (including Video / Visual Question Answer), but the video-based ones seem to be one
2.2 CVPR 2018
- 不完全统计有 15 篇(包括Video / Visual Question Answer),但是基于视频的好像就一篇
2.3 CVPR 2017
- 不完全统计有 9 篇(包括Video / Visual Question Answer),没有基于视频的
3.3 CVPR 2016
- 不完全统计有 8 篇(包括Video / Visual Question Answer),没有基于视频的,而且看起来是刚起步
3. ICCV
ICCV 全称 International Conference on Computer Vision, 中文名为国际计算机视觉大会,每两年在全世界范围内召开一次,录用率比较低,所以在业内评价较高,是三大CV顶会中公认级别最高的。
3.1 ICCV 2019
- 不完全统计有 5 篇(包括Video / Visual Question Answer)
3.2 ICCV 2017
- 不完全统计有 6 篇(包括Video / Visual Question Answer)
3.3 ICCV 2015
- 听名字感觉像是第一篇
论文题目 | 作者单位 |
---|---|
VQA: Visual Question Answering |