1:00, IMDSP-L2.1
DETECTION OF SLOW-MOTION REPLAY SEGMENTS IN SPORTS VIDEO FOR HIGHLIGHTS GENERATION
H. PAN, P. VAN BEEK, M. SEZAN
In this paper, we present a novel method for generating sports video summary highlights. Specifically, our method localizes semantically important events in sport programs by detecting slow motion replays of these events, and then generates highlights of these events at multiple levels. In our method, a hidden Markov model (HMM) is used to model slow motion replays, and an inference algorithm is introduced which computes the probability of a slow motion replay segment, and localizes the boundaries of the segment as well. An effective new feature is used in our HMM, based on a moving measure of the number of zero-crossings and the amplitudes of variations over time of video field differences. Furthermore, the method is capable of filtering out slow motion play segments in commercials. As compared with existing methods for video event detection, our method is more generic (i.e., domain independent), and has the ability to capture inherently important events.

1:20, IMDSP-L2.2
A ROBUST ALGORITHM FOR FUSING NOISY DEPTH ESTIMATES USING STOCHASTIC APPROXIMATION
A. ROY CHOWDHURY, R. CHELLAPPA
The problem of structure from motion (SFM) is to extract the three-dimensional model of a moving scene from a sequence of images. Most of the algorithms which work by fusing the two- frame depth estimates (observations) assume an underlying statistical model for the observations and do not evaluate the quality of the individual observations. However, in real scenarios, it is often difficult to justify the statistical assumptions. Also, outliers are present in any observation sequence and need to be identified and removed from the fusion algorithm. In this paper, we present a recursive fusion algorithm using Robbins-Monro stochastic approximation(RMSA) which takes care of both these problems to provide an estimate of the real depth of the scene point. The estimate converges to the true value asymptotically. We also propose a method to evaluate the importance of the successive observations by computing the Fisher information (FI) recursively. Though we apply our algorithm in the SFM problem by modeling of human face, it can be easily adopted to other data fusion applications.

1:40, IMDSP-L2.3
A NOVEL LINEAR TECHNIQUE TO ESTIMATE THE EPIPOLAR GEOMETRY
E. IZQUIERDO, V. GUERRA
A NOVEL LINEAR TECHNIQUE TO ESTIMATE THE EPIPOLAR GEOMETRY Ebroul Izquierdo(*) and Valia Guerra(**) (*) Department of Electronic Engineering Queen Mary, University of London London E1 4NS, United Kingdom (**) Group of Numerical Methods Institute of Math, Cybernetics and Physics Havana, Cuba ABSTRACT The accurate reconstruction of the 3D scene structure from two different projections and the estimation of the camera scene geometry is of paramount importance in many computer vision tasks. Most of the information about the camera-scene geometry is encapsulated in the Fundamental Matrix. Estimating the Fundamental Matrix has been an object of research for many years and continues to be a challenging task in current computer vision systems. While nonlinear iterative approaches have been successful in dealing with the high instability of the underlying problem, their inherent large workload makes these approaches inappropriate for real-time applications. In this paper practical aspects of highly efficient linear methods are studied and a novel low-cost and accurate linear algorithm is introduced. The performance of the proposed approach is assessed by several experiments on real images.

2:00, IMDSP-L2.4
A PROJECTION METHOD TO GENERATE ANAGLYPH STEREO IMAGES
E. DUBOIS
An anaglyph image allows the perception of depth when observed through colored glasses such as the familiar red/blue glasses. Although the method is very old, the techniques used to generate anaglyph images are very empirical. This paper describes a projection method to generate anaglyph stereoscopic images using the spectral absorption curves of the glasses, the spectral density functions of the display primaries and the colorimentric properties of the human observer. Sample images generated with this method can be found at the URL http://www.site.uottawa.ca/~edubois/icassp00/

2:20, IMDSP-L2.5
OPTIMAL HISTOGRAM MODIFICATION WITH MSE METRIC
M. MESE, P. VAIDYANATHAN
In this paper we propose a method to modify the histogram of a signal to a desired specific histogram. Traditionally, points having the same value in the input signal are all mapped to same value in the output signal. Hence, the desired histogram can only be approximated. Here we formulate our problem as finding a transformation such that the error between the input and output signal is minimized and the output signal has the desired histogram. It turns out that this problem is equivalent to an integer linear programming problem. This method might be specifically useful for histogram based watermarking and compression.

2:40, IMDSP-L2.6
BLUR ESTIMATION IN LIMITED-CONTROL ENVIRONMENTS
J. PRICE, T. GEE, K. TOBIN
In this paper, we propose a method to estimate the blur of a fixed imaging system, without control of camera position or lighting, using an inexpensive target. Such a method is applicable, for example, in the restoration of surveillance imagery where the imaging system is available, but with only limited-control of the imaging conditions. We extend a previously proposed parametric blur model and maximum likelihood technique to estimate a more general family of blur functions. The requirements for an appropriate characterization target are also discussed. Experimental results with artificial and real data are presented to validate the proposed approach.