Estimation of Goodness of Image Retrieval

2014 Fiscal Year Final Research Report

• Project/Area Number: 25540076
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: National Institute of Informatics
• Principal Investigator: SATOH Shin'ichi 国立情報学研究所, コンテンツ科学研究系, 教授 (90249938)
• Project Period (FY): 2013-04-01 – 2015-03-31
• Keywords: 画像データベース / 画像検索 / 画像レコメンデーション / 転置索引 / 分枝限定法

Outline of Final Research Achievements

We studied the estimation of goodness of image retrieval, namely, a method to judge if a given query returns good image retrieval results, without performing actual retrieval. On top of this, we implemented image recommendation system which can recommend good query with given image database. We tested the system with the database contains 1 million images and confirmed that our method can effectively recommend good query region with scalable processing.

Free Research Field

マルチメディア検索