Engineering Imagination -- Inferring past frames using massive video instances

2014 Fiscal Year Final Research Report

• Project/Area Number: 25540070
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: Kyushu University
• Principal Investigator: SEIICHI Uchida 九州大学, システム情報科学研究科(研究院, 教授 (70315125)
• Project Period (FY): 2013-04-01 – 2015-03-31
• Keywords: パターン認識 / 映像解析 / 映像合成 / 時系列処理 / 大規模データ

Outline of Final Research Achievements

The aim of this research is to mimic the imagination skill of human-being by computer. Human can imagine the process that an object falls down and bounds at the floor just by watching a final image where the object lies on the floor. One assumption for this imagination skill is that we refer to similar instances that we have already seen before. Based on the assumption, we prepare a large instance data set and realize a method to estimate the past frame images by referring instances appropriate for the individual frame images.

Free Research Field

画像情報学