Research on removal of real objects from images in video see-through augmented reality

2014 Fiscal Year Final Research Report

• Project/Area Number: 24700118
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Media informatics/Database
• Research Institution: Nara Institute of Science and Technology
• Principal Investigator: KAWAI Norihiko 奈良先端科学技術大学院大学, 情報科学研究科, 助教 (30610670)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 隠消現実感 / 画像修復 / 拡張現実感 / 物体除去 / 背景推定

Outline of Final Research Achievements

This study developed diminished reality methods for visually removing real objects from video images in real time in video see-through augmented reality. Specifically, we developed two methods for the cases when target objects are augmented reality markers and general objects. For augmented reality markers, we achieved real-time marker removal by automatically detecting AR markers and generating background images behind markers using image inpainting, and overlaying the generated images on the marker regions with geometric and photometric corrections. For general objects, we first estimated camera poses and geometry behind a target object on the basis of a SLAM (Simultaneous Localization and Mapping) technique. We then achieved real-time removal of general objects for a 3D scene by generating background images using image inpainting and overlaying the generated images with geometric and photometric corrections using the estimated camera poses and background geometry.

Free Research Field

画像・映像処理、複合現実感