Image reconstruction with the adaptively learned image model

2014 Fiscal Year Final Research Report

• Project/Area Number: 23700194
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: TANAKA Masayuki 東京工業大学, 理工学研究科, 准教授 (60401543)
• Project Period (FY): 2011-04-28 – 2015-03-31
• Keywords: 画像処理

Outline of Final Research Achievements

In this project, the image model which focuses on the image patch is proposed. In the existing algorithm, the model is learned with a training dataset. However, if the statistical properties of the training dataset is different from a target scene, an algorithm using that model cannot reach the good performance. In the proposed algorithm, the model is adaptively learned with the input image. The proposed algorithm is applied to the image denoising and the image super-resolution. The experimental results demonstrate that the proposed algorithm improves the performance of the image denoising and the image super-resolution.

Free Research Field

画像処理