A Survey of Inverse Rendering Problems

Inverse rendering problems usually represent extremely complex and costly processes, but their importance in many research areas is well known. In particular, they are of extreme importance in lighting engineering, where potentially costly mistakes usually make it unfeasible to test design decisions on a model. In this survey we present the main ideas behind these kinds of problems, characterize them, and summarize work developed in the area, revealing problems that remain unsolved and possible areas of further research. ACM CSS: I.3.6 Computer Graphics Methodology and Techniques I.3.7 Computer Graphics—Three‐Dimensional Graphics and Realism I.4.1 Image Processing and Computer Vision Digitization and Image Capture I.4.7 Image Processing and Computer Vision Feature Measurement I.4.8 Image Processing and Computer Vision Scene Analysis     

Surface and normal ensembles for surface reconstruction

The majority of the existing techniques for surface reconstruction and the closely related problem of normal reconstruction are deterministic. Their main advantages are the speed and, given a reasonably good initial input, the high quality of the reconstructed surfaces. Nevertheless, their deterministic nature may hinder them from effectively handling incomplete data with noise and outliers. An ensemble is a statistical technique which can improve the performance of deterministic algorithms by putting them into a statistics based probabilistic setting. In this paper, we study the suitability of ensembles in normal and surface reconstruction. We experimented with a widely used normal reconstruction technique [Hoppe H, DeRose T, Duchamp T, McDonald J, Stuetzle W. Surface reconstruction from unorganized points. Computer Graphics 1992;71-8] and Multi-level Partitions of Unity implicits for surface reconstruction [Ohtake Y, Belyaev A, Alexa M, Turk G, Seidel H-P. Multi-level partition of unity implicits. ACM Transactions on Graphics 2003;22(3):463-70], showing that normal and surface ensembles can successfully be combined to handle noisy point sets.     

Unifying statistical texture classification frameworks

The objective of this paper is to examine statistical approaches to the classification of textured materials from a single image obtained under unknown viewpoint and illumination. The approaches investigated here are based on the joint probability distribution of filter responses.

We review previous work based on this formulation and make two observations. First, we show that there is a correspondence between the two common representations of filter outputs—textons and binned histograms. Second, we show that two classification methodologies, nearest neighbour matching and Bayesian classification, are equivalent for particular choices of the distance measure. We describe the pros and cons of these alternative representations and distance measures, and illustrate the discussion by classifying all the materials in the Columbia-Utrecht (CUReT) texture database.

These equivalences allow us to perform direct comparisons between the texton frequency matching framework, best exemplified by the classifiers of Leung and Malik [Int. J. Comput. Vis. 43 (2001) 29], Cula and Dana [Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2001) 1041], and Varma and Zisserman [Proceedings of the Seventh European Conference on Computer Vision 3 (2002) 255], and the Bayesian framework most closely represented by the work of Konishi and Yuille [Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2000) 125].     

Industrial geometry: recent advances and applications in CAD

Industrial Geometry aims at unifying existing and developing new methods and algorithms for a variety of application areas with a strong geometric component. These include CAD, CAM, Geometric Modelling, Robotics, Computer Vision and Image Processing, Computer Graphics and Scientific Visualization. In this paper, Industrial Geometry is illustrated via the fruitful interplay of the areas indicated above in the context of novel solutions of CAD related, geometric optimization problems involving distance functions: approximation with general B-spline curves and surfaces or with subdivision surfaces, approximation with special surfaces for applications in architecture or manufacturing, approximate conversion from implicit to parametric (NURBS) representation, and registration problems for industrial inspection and 3D model generation from measurement data. Moreover, we describe a ‘feature sensitive’ metric on surfaces, whose definition relies on the concept of an image manifold, introduced into Computer Vision and Image Processing by Kimmel, Malladi and Sochen. This metric is sensitive to features such as smoothed edges, which are characterized by a significant deviation of the two principal curvatures. We illustrate its applications at hand of feature sensitive curve design on surfaces and local neighborhood definition and region growing as an aid in the segmentation process for reverse engineering of geometric objects.     

An efficient method for geo-referenced video mosaicing for environmental monitoring

Environmental monitoring applications require seamless registration of optical data into large area mosaics that are geographically referenced to the world frame. Using frame-by-frame image registration alone, we can obtain seamless mosaics, but it will not exhibit geographical accuracy due to frame-to-frame error accumulation. On the other hand, the 3D geo-data from GPS, a laser profiler, an INS system provides a globally correct track of the motion without error propagation. However, the inherent (absolute) errors in the instrumentation are large for seamless mosaicing. The paper describes an effective two-track method for combining two different sources of data to achieve a seamless and geo-referenced mosaic, without 3D reconstruction or complex global registration. Experiments with real airborne video images show that the proposed algorithms are practical in important environmental applications. Zhigang Zhu received his B.E., M.E. and Ph.D. degrees, all in computer science from Tsinghua University, Beijing, in 1988, 1991 and 1997, respectively. He is currently an associate professor in the Department of Computer Science, the City College of the City University of New York. Previously, he was an associate professor at Tsinghua University, and a senior research fellow at the University of Massachusetts, Amherst. His research interests include 3D computer vision, HCI, virtual/augmented reality, video representation, and various applications in education, environment, robotics, surveillance and transportation. He has published over 90 technical papers in the related fields. He is a member of IEEE and ACM. Edward M. Riseman received his B.S. degree from Clarkson College of Technology in 1964 and his M.S. and Ph.D. degrees in electrical engineering from Cornell University in 1966 and 1969, respectively. He joined the Computer Science Department at UMass-Amherst as assistant professor in 1969, has been a professor since 1978, and served as chairman of the department from 1981 to 1985. Professor Riseman has conducted research in computer vision, artificial intelligence, learning, and pattern recognition, and has more than 200 publications. He has co-directed the Computer Vision Laboratory since its inception in 1975. Professor Riseman has been on the editorial boards of Computer Vision and Image Understanding (CVIU) from 1992 to 1997 and of the International Journal of Computer Vision (IJCV) from 1987 to the present. He is a senior member of IEEE, and a fellow of AAAI. Allen R. Hanson received his B.S. degree from Clarkson College of Technology in 1964 and his M.S. and Ph.D. degrees in electrical engineering from Cornell University in 1966 and 1969, respectively. He joined the Computer Science Department at UMass-Amherst as an associate professor in 1981, and has been a professor there since 1989. Professor Hanson has conducted research in computer vision, artificial intelligence, learning, and pattern recognition, and has more than 150 publications. He is co-director of the Computer Vision Laboratory at UMass-Amherst, and has been on the editorial boards of the following journals: Computer Vision, Graphics and Image Processing 1983–1990, Computer Vision, Graphics, and Image Processing—Image Understanding 1991–1994, and Computer Vision and Image Understanding 1995–present. Howard Schultz received a M.S. degree in physics from UCLA in 1974 and a Ph.D. in physical oceanography from the University of Michigan in 1982. Currently, he is a senior research fellow with the Computer Science Department at the University of Massachusetts, Amherst. His research interests include quantitative methods for image understanding and remote sensing. The current focus of his research activities are on developing automatic techniques for generating complex, 3D models from sequences of images. This research has found application in a variety of programs including real-time terrain modeling and video aided navigation. He is a member of the IEEE, the American Geophysical Union, and the American Society of Photogrammetry and Remote Sensing.