Research Projects:

Near Light Correction for Image Relighting and 3D Shape Recovery
Xiang Huang, Marc Walton, Greg Bearman and Oliver Cossairt, to appear in Digital Heritage 2015 (long paper)
[Project Page] [PDF]
We propose a near-light illumination model for image relighting and 3D shape recovery. We correct the near-light distance effectto provide much more uniformly lit images that yield more appealing image for relighting applications. Furthermore, we use our near-light model for more accurate photometric stereo calculations of surface normals, eliminating the “potato-chip” shaped surface reconstruction error that results from violating the far-light assumption.
We verify our model with both free-form capture using hand-held flash as the illumination source, and capture using LED lights mounted on a dome shaped surface.



Surface Shape Studies of the Art of Paul Gauguin
Oliver Cossairt, Xiang Huang, Nathan Matsuda, Harriet Stratis, Mary Broadway, Jack Tumblin, Greg Bearman, Dale Kronkright, Eric Doehne, Aggelos Katsaggelos and Marc Walton, to appear in Digital Heritage 2015 (long paper)
[Project Page] [PDF]
Starting in the 1890s the artist Paul Gauguin (1848- 1903) created a series of monoprints and drawings using techniques that are not entirely understood. To better understand the artists production methods, photometric stereo was used to assess the surface shape of a number of these graphic works that are now housed at the Art Institute of Chicago. Photometric stereo uses multiple images of Gauguins graphic works captured from a fixed camera position, lit from various different directions to create an interactive composite image that reveals textural characteristics. These active images reveal details on the sequence of inks applied to the surfaces of the prints and blind incisions in the paper substrate that help resolve longstanding art historical questions about the evolution of Gauguins printing techniques. Our study promotes the use of photometric stereo to capitalize on the increasingly popularity of Reflectance Transformation Imaging (RTI) among conservators in the world’s leading museums.

Dictionary Learning based Color Demosaicing for Plenoptic Cameras
Xiang Huang and Oliver Cossairt,CCD'2014
[Project Page] [PDF]
We propose a dictionary learning based demosaicing algorithm that recovers a full-color light field from a captured plenoptic image using sparse optimization.


What Characterizes a Shadow Boundary under the Sun and Sky?
Xiang Huang, Gang Hua, Jack Tumblin and Lance Williams, ICCV2011
Paper (PDF)
Poster (PDF)

   Despite decades of study, robust shadow detection remains difficult, especially within a single color image. We describe a new approach to detect shadow boundaries in images of outdoor scenes lit only by the sun and sky. The method first extracts visual features of candidate edges that are motivated by physical models of illumination and occluders. We feed these features into a Support Vector Machine (SVM) that was trained to discriminate between most-likely shadow-edge candidates and less-likely ones. Finally, we connect edges to help reject non-shadow edge candidates, and to encourage closed, connected shadow boundaries. On benchmark shadow-edge data sets, our method showed substantial improvements when compared to other recent shadow-detection methods based on statistical learning.

Sensing Increased Image Resolution Using Aperture Masks
Ankit Mohan, Xiang Huang, Ramesh Raskar and Jack Tumblin, CVPR08 Paper (PDF)
Slides (PDF)
More details...

Discussions on Nuit Blanche

   We investigated a technique to construct increased-resolution images
from multiple photos taken without moving the camera or the sensor. Like other super-resolution techniques, we capture and merge multiple images, but instead of moving the camera sensor by sub-pixel distances for each image, we change masks in the lens aperture and slightly defocus the lens. The resulting capture system is simpler and tolerates modest mask registration errors well.

Deep Shadows in a Shallow Box
Xiang Huang, Ankit Mohan and Jack Tumblin
SPIE Electronic Imaging Conf. 2008
Paper (PDF)  Slides (PDF)

    We present a fast, low-cost technique to gather high-contrast `relightable' photographs of desktop-sized objects. By removing the ambient light computationally, we generate high-contrast deep shallow images from originally captured low-contrast shallow shadow images.

Multiple Description Lattice Vector Quantization (MDLVQ)
Xiang Huang and Xiaolin Wu
DCC'06 Paper (PDF)  Journal Draft (PDF)   Thesis (PDF)

   Multiple description coding is motivated by cooperate and distributed source coding for packet lossy networks. MDLVQ is an e ective multiple description coding scheme. We proposed an optimal linear-time MDLVQ index assignment algorithm.


Last update: Jun 10, 2015