Home / Publications


IEEE Computer Graphics and Applications 2014
Multispectral Photometric Stereo for Acquiring High-Fidelity Surface Normals
  Giljoo Nam Min H. Kim  
  Schematic diagram of our multispectral photometric stereo that removes interreflection effectively. (a) shows the measurement setup and captured image of the target object (L-shaped in 90◦). The inner faces of the object present interreflection along with direct reflection. (b) presents the radiometric power distribution (captured by a multispectral camera) over the seven points on the left-hand-side surface with interreflection. (c) compares the reconstructed 3D geometries using ordinary photometric stereo with interreflection and our method that removes interreflection using multispectral photometric stereo.  
  An advanced imaging technique of multispectral imaging has become more accessible as a physically-meaningful image-based measurement tool, and photometric stereo has been commonly practiced for digitizing a 3D shape with simplicity for more than three decades. However, these two imaging techniques have rarely been combined as a 3D imaging application yet. Reconstructing the shape of a 3D object using photometric stereo is still challenging due to the optical phenomena such as indirect illumination, specular reflection, self shadow. In addition, removing interreflection in photometric stereo is a traditional chicken-and-egg problem as we need to account for interreflection without knowing geometry. In this paper, we present a novel multispectral photometric stereo method that allows us to remove interreflection on diffuse materials using multispectral reflectance information. Our proposed method can be easily integrated into an existing photometric stereo system by simply substituting the current camera with a multispectral camera, as our method does not rely on additional structured or colored lights. We demonstrate several benefits of our multispectral photometric stereo method such as removing interreflection and reconstructing the 3D shapes of objects to a high accuracy.
  author  = {Giljoo Nam and Min H. Kim},
  title   = {Multispectral Photometric Stereo for Acquiring 
            High-Fidelity Surface Normals},
  journal = {IEEE Computer Graphics and Applications},
  year    = {2014},
  volume  = {34},
  number  = {6},
  pages   = {57--68},
  doi     = "10.1109/MCG.2014.108",
Prepress paper:
PDF (2.5MB)
icon Video:
MP4 (4.1MB)
icon IEEE DL

Hosted by Visual Computing Laboratory, School of Computing, KAIST.