Computer Graphics and Visualization › Publications › Indoor Scene Reconstruction Using Near-Light Photometric Stereo

IEEE Trans Image Process

Indoor Scene Reconstruction Using Near-Light Photometric Stereo

Jingtang Liao, Bert Buchholz, Jean-Marc Thiery, Pablo Bauszat, and Elmar Eisemann

Comparison (normal map) of our approach with the two state-of-the-art near-light PS algorithms from Ahmad et al. [4] and Mecca et al. [32] for the MONKEY scene. Please note that shadows and discontinuities in our input makes the data already unsuitable for these algorithms, hence it is obvious that their reconstructions fail for most parts.

Publisher Paper Cite More

We propose a novel framework for photometric stereo (PS) under low-light conditions using uncalibrated nearlight illumination. It operates on free-form video sequences captured with a minimalistic and affordable setup. We address issues such as albedo variations, shadowing, perspective projections, and camera noise. Our method uses specular spheres detected with a perspective-correcting Hough transform to robustly triangulate light positions in the presence of outliers via a least-squares approach. Furthermore, we propose an iterative reweighting scheme in combination with an p-norm minimizer to robustly solve the calibrated near-light PS problem. In contrast to other approaches, our framework reconstructs depth, albedo (relative to light source intensity), and normals simultaneously and is demonstrated on synthetic and real-world scenes.

More Information

Publisher Paper preprint

Citation

Jingtang Liao, Bert Buchholz, Jean-Marc Thiery, Pablo Bauszat, and Elmar Eisemann, Indoor Scene Reconstruction Using Near-Light Photometric Stereo, IEEE Trans Image Process, 26, pp. 1089–1101, 2017.

BibTex 

@article{bib:liao:2017,
    author       = { Liao, Jingtang and Buchholz, Bert and Thiery, Jean-Marc and Bauszat, Pablo and Eisemann, Elmar },    
    title        = { Indoor Scene Reconstruction Using Near-Light Photometric Stereo },
    journal      = { IEEE Trans Image Process },
    volume       = { 26 },
    year         = { 2017 },
    pages        = { 1089--1101 },
    doi          = { 10.1109/TIP.2016.2636661 },
    dblp         = { journals/tip/LiaoBTBE17 },
    url          = { https://publications.graphics.tudelft.nl/papers/239 },
}