Image Reconstruction from Bag of Visual Words using Large-scale Image Data

2015 Fiscal Year Final Research Report

• Project/Area Number: 26540079
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: The University of Tokyo
• Principal Investigator: Harada Tatsuya 東京大学, 情報理工学(系)研究科, 教授 (60345113)
• Project Period (FY): 2014-04-01 – 2016-03-31
• Keywords: コンピュータビジョン / 機械学習

Outline of Final Research Achievements

The objective of this study is to reconstruct images from Bag-of-Visual-Words (BoVW). BoVW is defined here as a histogram of quantized local descriptors extracted densely on a regular grid at a single scale. This task is challenging for two reasons: 1) BoVW includes quantization errors. 2) BoVW lacks spatial information of local descriptors. To tackle this difficult task, we use a large-scale image database to estimate the spatial arrangement of local descriptors. Then this task creates a jigsaw puzzle problem with adjacency and global location costs of visual words. Solving this optimization problem is also challenging because it is known as an NP-Hard problem. We propose a heuristic but efficient method to optimize it. To underscore the effectiveness of our method, we apply it to BoVWs extracted from about 100 different categories and demonstrate that it can reconstruct the original images.

Free Research Field

知能機械情報学