VCLab

 

RESEARCH AREAS   PEOPLE   PUBLICATIONS   COURSES   ABOUT US
Home / Publications

indicator

ACM SIGGRAPH Asia 2018 (Transactions on Graphics)

 
Practical SVBRDF Acquisition of 3D Objects with Unstructured Flash Photography
 
  Giljoo Nam Joo Ho Lee Diego Gutierrez* Min H. Kim  
  KAIST *Universidad de Zaragoza, I3A  
           
 
  We propose a practical method that enables us to capture spatially-varying BRDFs from unstructured flash photographs. It yields high-quality SVBRDFs, as well as detailed geometry of 3D objects, without relying on any expensive supporting hardware nor controlled illumination, using instead any hand-held conventional digital camera with a built-in flash. No input geometry is needed in our algorithm. Image (a) shows our acquisition setup, while (b)---(e) show example results of our reconstructions.  
     
     
   
     
   
  Abstract
   
  Capturing spatially-varying bidirectional reflectance distribution functions (SVBRDFs) of 3D objects with just a single, hand-held camera (such as an off-the-shelf smartphone or a DSLR camera) is a difficult, open problem. Previous works are either limited to planar geometry, or rely on previously scanned 3D geometry, thus limiting their practicality. There are several technical challenges that need to be overcome: First, the built-in flash of a camera is almost colocated with the lens, and at a fixed position; this severely hampers sampling procedures in the light-view space. Moreover, the near-field flash lights the object partially and unevenly. In terms of geometry, existing multiview stereo techniques assume diffuse reflectance only, which leads to overly smoothed 3D reconstructions, as we show in this paper. We present a simple yet powerful framework that removes the need for expensive, dedicated hardware, enabling practical acquisition of SVBRDF information from real-world, 3D objects with a single, off-the-shelf camera with a built-in flash. In addition, by removing the diffuse reflection assumption and leveraging instead such SVBRDF information, our method outputs high-quality 3D geometry reconstructions, including more accurate high-frequency details than state-of-the-art multiview stereo techniques. We formulate the joint reconstruction of SVBRDFs, shading normals, and 3D geometry as a multi-stage, iterative inverse-rendering reconstruction pipeline. Our method is also directly applicable to any existing multiview 3D reconstruction technique. We present results of captured objects with complex geometry and reflectance; we also validate our method numerically against other existing approaches that rely on dedicated hardware, additional sources of information, or both.
   
  BibTeX
 
  @Article{MobileSVBRDF:SIGA:2018,
  author  = {Giljoo Nam and Joo Ho Lee  
             and Diego Gutierrez and Min H. Kim},
  title   = {Practical SVBRDF Acquisition of 3D Objects 
             with Unstructured Flash Photography},
  journal = {ACM Transactions on Graphics (Proc. SIGGRAPH Asia 2018)},
  year    = {2018},
  volume  = {37},
  number  = {6},
  pages   = {267:1--12},
  doi     = "10.1145/3272127.3275017",
  url     = "http://dx.doi.org/10.1145/3272127.3275017",
  }              
   
   
   
icon
Preprint paper:
PDF (20MB)
icon
Supplemental document:
PDF (31MB)

SIGGRAPH Asia 2018 slides
PDF (TBA)
www ACM
Digital Library

www Press Release:
EurekAlert, ScienceDaily, E&T Magazine, VR Soldier, SciTech
 

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

KAIST