Projector Defocus Correction by Pre-Sharpened Image Projection

Abstract

We have developed projector defocus correction methods without implicit fiducial pattern projection. Most of existing projector image adjustment works, e.g., geometric adjustment and color adjustment, are constrained by hardware limitations of projector. Narrow depth-of-field results defocus blur with volumetric depth or moving screen object. When its resolution or dynamic range is not enough for the purpose, displayed image quality is significantly lost. For dynamic projection systems, one of the most important things is defocus blur. Thus, in-focus projection technique is a good topic for the future improvements of projector.

Thanks to the recent development of image feature descriptor, representation, and matching and tracking works, projector image adjustment researches are going to remove structured light projection from their method. This trend is persuasive because structured light projection is a tool not a purpose. Even though we can expect imperceptible structured light to provide as same quality result as normal structured light does, embedding the pattern itself is undesirable additional process. Therefore, developing image matching based approaches can be a suitable choice for a system under dynamic scene. Since most projector-camera researchers carefully use projector to make the display in-focus, focal adjustment was previously less required. Furthermore, in-focus projection technique is a fresh research field. Therefore, there does not exist an in-focus projection method based on image matching method.

The potential application that the proposed defocus correction method can contribute is projector based application such as spatial augmented reality systems. One of the big limitations of such systems is projectors' shallow depth-of-field. We can solve the problem by integrating our method.

Publication

Please contact charmie hvrl.ics.keio.ac.jp if you want any pre-prints.

• [Oyamada 2009 PROCAMS] Yuji Oyamada and Hideo Saito, "Blind Deconvolution Based Projector Defocus Removing with Uncalibrated Projector-Camera Pair," IEEE International Workshop on Projector-Camera Systems (PROCAMS), 2009, [paper, bib]
• [Oyamada 2008 ACIVS] Yuji Oyamada and Hideo Saito, "Defocus Blur Correcting Projector-Camera System," Advanced Concepts for Intelligent Vision Systems (ACIVS), 2008, [paper, bib]
• [Oyamada 2008 WSCG] Yuji Oyamada and Hideo Saito, "Estimation of Projector Defocus Blur by Extracting Texture Rich Regionin Projection Image," International Conference in Central Europe on Computer Graphics,Visualization and Computer Vision (WSCG), 2008, [paper, bib]
• [Oyamada 2007 PROCAMS] Yuji Oyamada and Hideo Saito, "Focal Pre-Correction of Projected Image for Deblurring Screen Image," IEEE International Workshop on Projector-Camera Systems (PROCAMS), 2007, [paper, bib]
• [Oyamada 2007 FCV] Yuji Oyamada and Hideo Saito, "Pre-Correction of Projected Images for Deblurring on Projector Screen," Japan-Korea Joint Workshop on Frontiers of Computer Vision (FCV), 2007, [paper, bib]

Acknowledges

This work was partially supported in part by a Grant-in-Aid for the Global Center of Excellence for High-Level Global Cooperation for Leading-Edge Platform on Access Spaces from the Ministry of Education, Culture, Sport, Science, and Technology in Japan.