Ravi Ramamoorth: “Sampling and Reconstruction of High-Dimensional Visual Appearance”

October 24, 2013 | USC ICT

Speaker: Ravi Ramamoorth

Abstract: Many problems in computer graphics and computer vision involve high-dimensional 3D-8D visual datasets. Real-time image synthesis with changing lighting and view is often accomplished by pre-computing the 6D light transport function (2 dimensions each for spatial position, incident lighting and viewing direction). Realistic image synthesis also often involves acquisition of appearance data from real-world objects; a BRDF (Bi-Directional Reflection Distribution Function) that measures the scattering of light at a single surface location is 4D and spatial variation and subsurface scattering involve 6D-8D functions. Volumetric appearance datasets are increasingly used for realism and can involve billions of voxels in three dimensions. In computer vision, problems like lighting insensitive facial recognition similarly involve understanding the space of appearance variation across lighting and view.

Since hundreds of samples may be required in each dimension, and the total size is exponential in the dimensionality, brute force acquisition or precomputation is often not even feasible. In this talk, we describe a signal-processing approach that exploits the coherence, sparsity and inherent low-dimensionality of the visual data, to derive novel efficient sampling and reconstruction algorithms. We describe a variety of new computational methods and applications, from affine wavelet transforms for real-time rendering with area lights, to space-time and space-angle frequency analysis for motion blur and global illumination, to compressive light transport acquisition. In computer vision, we introduce a new framework of differential photometric reconstruction to tame the complexity of real-world reflectance functions. The results point toward a unified sampling theory applicable to many areas of signal processing, computer graphics and computer vision.

Bio: Ravi Ramamoorthi is an Associate Professor of Electrical Engineering and Computer Science at the University of California, Berkeley since Jan 2009. Earlier, he received his BS, MS degrees from Caltech and his PhD from Stanford University in 2002, after which he joined the faculty of the computer science department at Columbia University. He is interested in many areas of computer graphics, computer vision and signal-processing, having published more than 90 papers, including 45 in ACM SIGGRAPH/Transactions on Graphics. He has received a number of research awards, including the 2007 ACM SIGGRAPH Significant New Researcher Award for his work in computer graphics, and the 2008 White House US Presidential Early Career Award for Scientists and Engineers for his work on physics-based computer vision. He was also awarded the NSF Career Award (2005), Sloan Fellowship (2005), ONR Young Investigator Award (2007) and Okawa Foundation Research Grant (2011).
Many of his contributions, such as spherical harmonic lighting and importance sampling, are widely used in industry. He has also been a leader in education, teaching the first open online course in computer graphics with total enrollments to date of 50,000 students, and he has advised more than 20 postdoctoral, Ph.D. and MS students. Learn more at