ICASSP '98 Main Page

General Information

Conference Schedule

Technical Program

    Overview
    50th Annivary Events
    Plenary Sessions
    Special Sessions
    Tutorials
    Technical Sessions

	By Date
		May 12, Tue
		May 13, Wed
		May 14, Thur
		May 15, Fri

By Category
	AE	ANNIV
	COMM	DSP
	IMDSP	MMSP
	NNSP	PLEN
	SP	SPEC
	SSAP	UA
	VLSI

By Author
	A	B	C	D	E
	F	G	H	I	J
	K	L	M	N	O
	P	Q	R	S	T
	U	V	W	X	Y
	Z

Invited Speakers

Registration

Exhibits

Social Events

Coming to Seattle

Satellite Events

Call for Papers/
Author's Kit

Future Conferences

Help

Abstract - IMDSP2

IMDSP2.1	Representation and Estimation of Motion Using a Dictionary of Models D. Lauzon, E. Dubois (INRS-Telecommunications, Canada) This paper presents a novel method for representing motion information based on a Dictionary of Motion Models and a Tag Image which indicates which motion model is used at any given image position. Each model is composed of low-order polynomial-based motion fields. The motion in most sequences can be adequately represented by a very small number of such motion models. We further present an efficient way of estimating and coding this representation. Comparative results are presented which indicate a performance superior to that of motion representations found in classical block-based codecs.
IMDSP2.2	Global Motion Estimation and Robust Regression for Video Coding K. Zhang, J. Kittler (CVSSP, University of Surrey, UK) In the H.263 Version 2 (H.263+) coding standard, the global motion compensation can be introduced by using Reference Picture Resampling (Annex. P) syntax. Such an application requires that the global motion parameters be estimated automatically. In this paper, we propose a global motion estimation algorithm based on the Taylor Expansion Equation and robust regression technique using probabilistic thresholding. The experimental results confirm that the proposed algorithm can improve both coding efficiency and the quality of motion compensation on sequences involving camera movement.
IMDSP2.3	Parametric Motion Modeling Based on Trilinear Constraints for Object-Based Video Compression Z. Sun, A. Tekalp (University of Rochester, USA) We propose a new parametric motion model based on the so-called ``trifocal tensor" representation, which captures rigid 3D motion of static scenes with a depth of field. The proposed parametric representation, called the trilinear model, is superior to other forms such as translational, affine, perspective, and bilinear models, because it can implicitly encode the depth of the scene and 3D motion of the scene/camera under perspective projection unlike others. A video object can thus be represented by its first VOP, a set of trifocal sensors and the corresponding prediction residues. Motion estimation and compensation based on the new parametric model are incorporated into the MPEG-4 Video Verification Model to compare its efficacy for object-based video compression with the state-of-the-art motion compensation methods. Experimental results are provided to demonstrate the performance of the trilinear model for object-based video compression.
IMDSP2.4	Adaptive Thresholding for Detection of Nonsignificant Vectors in Noisy Image Sequences L. Martel, A. Zaccarin (Laval University, Canada) In noisy image sequences, block matching motion estimation generates erroneous motion vectors since the algorithm tries to correlate noise. We present an adaptive threshold test to detect blocks for which only nonsignificant motion vectors can be estimated. Vectors of these blocks are then assigned the zero vector before any block motion estimation is performed. By nonsignificant, we refer to motion vectors of non moving areas as well as vectors of moving areas for which the noise level is too high to allow a good estimation of the motion. The detection of these vectors reduces the computational complexity of the BMA and the entropy of the motion field. The algorithm is embedded in a hierarchical BMA and takes advantage of their different spectral characteristics to discriminate between the frame difference energy due to noise and due to motion. The algorithm is also efficient for low noise sequences where it can be used to initialize a segmentation of moving objects from the background.
IMDSP2.5	Computing Optical Flow for Motion Images Sequences by Applying Multi-constraints to Multi-locations C. Yang, S. Oe (University of Tokushima, Japan) Computing optical flow is one of the most fundamental problem to the motion image analysis. Many methods have been proposed for computing optical flow, among them gradient-based methods are the most well-known and most used. In the paper, a new gradient-based method for the computation of optical flow was proposed. In this method, optical flow was computed by minimizing a weighted least-squares error estimator for a constant motion vector model in a local spatial neighborhood, where the weight of each image location in the neighborhood was determined by its multiple constraints. Several experiments on real and synthetic image sequences have been carried out to verify the efficacy and the reliability of the new method.
IMDSP2.6	Active Mesh Reconstruction of Block-Based Motion Information X. Marichal, B. Macq (Lab de Telecommunication & Teledetection, Belgium) This paper proposes an asymmetric scheme for motion estimation/compensation. While the estimation is performed with a classical Block Matching Algorithm, the motion information is decoded by using an active mesh in order to implement the compensation stage. A mesh is positioned by taking into account the relevant spatial information of the image to be compensated and is used afterwards to reconstruct the motion information. Two main issues have to be addressed for conducting such a motion compensation technique: i) how to optimally design an active mesh, ii) how to reverse and interpolate a backward motion field estimated on an a priori grid of fixed-size blocks so as to determine a forward motion field on variable size triangular patches. While proposing a solution to these two problems, particular attention is paid to the computational burden. Such a scheme opens the possibility for added manipulation functionalities because of mesh capabilities.
IMDSP2.7	Automatic Fitting and Tracking of Facial Features in Head-And-Shoulders Sequences P. Antoszczyszyn, J. Hannah, P. Grant (The University of Edinburgh, Scotland, UK) Model-based video coding requires the application of both image processing and machine vision techniques for proper fitting of the semantic model and its subsequent tracking throughout the rest of the sequence of a certain type (e.g. 'head-and-shoulders' or 'head-only'). A method of automatic semantic wire-frame fitting and tracking based on principal component analysis using an independent reference data-base of facial images is presented. The method has been tested on widely used 'head-and-shoulders' video sequences with very good results. It was possible to accurately retrieve the position of the desired facial features in all cases. The position of the facial features in initial frames was subsequently used in automatic tracking. Experimental results are presented as a part of this contribution. Compressed movies illustrating these results can be viewed from our Internet site http://www.ee.ed.ac.uk/~plma/.
IMDSP2.8	Fast Rate-Constrained N-Step Search Algorithm for Motion Estimation M. Coban, R. Mersereau (Georgia Institute of Technology, USA) A fast N-step search algorithm for rate-constrained motion estimation is presented. The motion vectors are selected from a search window based on a rate-distortion criterion by successively eliminating the search positions at each step. The performance of the proposed algorithm is identical to the performance of the conventional rate-constrained N-step search algorithm, with considerable reduction in computation. Computational savings increase in parallel with the increases in the rate constraint and the number of steps.

< Previous Abstract - IMDSP1

IMDSP3 - Next Abstract >