News

  • 10/2018:   Our paper on single-senosr RGB-NIR imaging has been accepted for IEEE Sensors Journal. [Project]
  • 10/2018:   We released our 59-band (420nm-1000nm) hyperspectral image dataset. [Link]
  • 07/2018:   We presented a paper on remote heart rate measurement at EMBC2018. [Project]
  • 12/2017:   Our paper on color and multispectral demosaicking has been accepted for Sensors. [Project]
  • 11/2017:   Our paper got IEEE signal processing society (SPS) Japan best paper award.
  • 09/2017:   Yusuke Monno gave a sourt course on spectral filter arrays at CIC25. [Slides]
  • 06/2017:   We released our 31-band (420nm-720nm) multispectral image dataset. [Link]
  • 06/2017:   We released our 5-band multispectral RAW image dataset. [Link]
  • 12/2016:   We released our 5-band multispectral image dataset and demosaicking codes. [Link]
  • 06/2016:   We presented a live demonstration of our prototype RGB-NIR imaging system at CVPR2016. [Project]
  • 06/2016:   Our prototype RGB-NIR imaging system appears in many online news. [List]
  • 06/2016:   We made a press release about our prototype RGB-NIR imaging system. [Link]
  • 02/2016:   Yusuke Monno got Seiichi Tejima Doctoral Dissertation Award for this project.
  • 02/2016:   We gave a talk at Rader Interferometry Group, Stanford University, USA.
  • 02/2016:   Yusuke Monno got Inoue Research Award for Young Scientists for this project.
  • 11/2015:   Our paper has been accepted for Electronic Imaging 2016. [Project]
  • 11/2015:   Yusuke Monno got NF Foundation R&D Encouragement Award for this project.
  • 10/2015:   Our paper has been accepted for CPCV2015 (in conjunction with ICCV2015). [Project]
  • 10/2015:   We gave a talk at Natural Phenomena Simulation Group, University of Waterloo, Waterloo, Canada.
  • 05/2015:   Our paper has been accepted for IEEE Trans. on Image Processing. [Project]
  • 10/2014:   We gave a talk at ETIS, Universite de Cergy-Pontoise, Cergy, France.
  • 10/2014:   We presented one paper at ICIP2014. [Project]
  • 10/2014:   Page renewal.


Project Abstract

In this project, our goal is to develop a practical one-shot multispectral imaging system using a single image sensor with multispectral filter array (MSFA) and demosaicking, which significantly extends a range of computer vision and image processing applications.



Development of prototype RGB-NIR imaging system [Project]

  We have developed a prototype RGB-NIR imaging system that can capture RGB and NIR images simultaneously in real time.

  • Single-Sensor RGB-NIR Imaging: High-Quality System Design and Prototype Implementation, IEEE Sensors Journal, 2019.

Development of 5-band multispectral camera prototype [Project]

  We built a practical one-shot 5-band camera prototype that can capture high-quality 5-band multispectral images at 30fps.

  • A Practical One-Shot Multispectral Imaging System Using a Single Image Sensor, IEEE TIP, 2015.

Effective multispectral-to-sRGB mapping pipeline

  We proposed a simple and effective multispectral-to-sRGB mapping pipeline for single-sensor multispectral imaging. [Project]

  • N-to-sRGB Mapping for Single-Sensor Multispectral Imaging, CPCV2015.

The design of a multispectral filter array and demosaicking

  We developed a novel multispectral filter array and a high-performance demosaicking algorithm for one-shot multispectral imaging.

  • Multispectral Demosaicking with Novel Guide Image Generation and Residual Interpolation, ICIP2014. [Project]
  • Multispectral Demosaicking Using Guided Filter, EI2012. [Project]
  • Multispectral Demosaicking Using Adaptive Kernel Upsampling, ICIP2011. [Project]

Optimization of spectral sensitivity functions

  We optimized spectral sensitivity functions for our proposed multispectral filter array.

  • Optimal Spectral Sensitivity Functions for a Single-Camera One-Shot Multispectral Imaging System, ICIP2012. [Project]

Spectral reflectance estimation from raw MSFA data

  We proposed a novel method that estimates spectral reflectance directly from raw MSFA data.

  • Direct Spatio-Spectral Datacube Reconstruction from RAW Data Using a Spatially Adaptive Spatio-Spectral Basis, EI2013. [Project]

Simultaneous capturing of RGB and additional band images

  We proposed a novel CFA, which we call "Hybrid CFA," and demosaicking algorithm for a joint capture of RGB and additional band (for example, NIR, UV, and narrow band) images.

  • Simultaneous Capturing of RGB and Additional Band Images Using Hybrid Color Filter Array, EI2014. [Project]



Remote Heart Rate Measurement from RGB-NIR Video


  We proposed a novel remote heart rate measuremen method using simultaneously recorded RGB and NIR videos, which is robust to various illumination conditions. [Project]

  • Remote heart rate measurement from RGB-NIR video based on spatial and spectral face patch selection, EMBC2018.
Thursday, 18-Jul-2019 22:45:56 JST
カウンター