Showing papers on "Motion analysis published in 2001"

Method and system for physical motion analysis

Abstract: An analysis system and method for providing athletic training and instruction by sensing different types of information, such as video, positional information, weight transfer information etc and synchronizing the information. The synchronized information is replayed for the user in a manner that enables simultaneous viewing of an athletic motion along with calculations and presentation of analysis information related to the athletic motion.

Hierarchical motion history images for recognizing human motion

Abstract: There has been increasing interest in computer analysis and recognition of human motion. Previously we presented an efficient real-time approach for representing human motion using a compact "motion history image" (MHI). Recognition was achieved by statistically matching moment-based features. To address previous problems related to global analysis and limited recognition, we present a hierarchical extension to the original MHI framework to compute dense (local) motion flow directly from the MHI. A hierarchical partitioning of motions by speed in an MHI pyramid enables efficient calculation of image motions using fixed-size gradient operators. To characterize the resulting motion field, a polar histogram of motion orientations is described. The hierarchical MHI approach remains a computationally inexpensive method for analysis of human motions.

Interpolation of a sequence of images using motion analysis

Abstract: Two images are analyzed to compute a set of motion vectors that describes motion between the first and second images. A motion vector is computed for each pixel in an image at a time between the first and second images. This set of motion vectors may be defined at any time between the first and second images, such as the midpoint. The motion vectors may be computed using any of several techniques. An example technique is based on the constant brightness constraint, also referred to as optical flow. Each vector is specified at a pixel center in an image defined at the time between the first and second images. The vectors may point to points in the first and second images that are not on pixel centers. The motion vectors are used to warp the first and second images to a point in time of an output image between the first and second images using a factor that represents the time between the first and second image at which the output image occurs. The warped images are then blended using this factor to obtain the output image at the desired point in time between the first and second images. The point in time at which the output image occurs may be different from the time at which the motion vectors are determined. The same motion vectors may be used to determine two or more output images at different times between the first and second images. The images may be warped using a technique in which many small triangles are defined in an image corresponding in time to the point in time between the first and second images at which the motion vectors are determined. A transform for each small triangle from the point in time at which the motion vectors are determined to the desired interpolated image time is determined, e.g., the triangle is warped using the motion vectors associated with its vertices. For each pixel in each triangle in the output image, corresponding points in the first and second images are determined, and the first and second images are spatially sampled at these points. These samples for each pixel are combined to produce a value for that pixel in the output image.

Multiple camera fusion for multi-object tracking

Abstract: We propose a distributed, real-time computing platform for tracking multiple interacting persons in motion. To overcome occlusion and articulated motion we use a multi-view implementation, where 2-D semantic features are independently tracked in each view and then collectively integrated using a Bayesian belief network with a topology that varies as a function of scene content and feature confidence. The network fuses observations from multiple cameras by resolving independency relationships and confidence levels within the graph, thereby producing the most likely vector of 3-D state estimates given the available data. We demonstrate the efficacy of the proposed system using a multi-view sequence of several people in motion. Our experiments suggest that, when compared with data fusion based on averaging, the proposed technique yields a noticeable improvement in tracking accuracy.

A subspace approach to layer extraction

Abstract: Representing images with layers has many important applications, such as video compression, motion analysis, and 3D scene analysis. This paper presents an approach to reliably extracting layers from images by taking advantages of the fact that homographies induced by planar patches in the scene form a low dimensional linear subspace. Layers in the input images will be mapped in the subspace, where it is proven that they form well-defined clusters and can be reliably identified by a simple mean-shift based clustering algorithm. Global optimality is achieved since all valid regions are simultaneously taken into account, and noise can be effectively reduced by enforcing the subspace constraint. Good layer descriptions are shown to be extracted in the experimental results.

Scene analysis with saccadic eye movements: Top-down and bottom-up modeling

Abstract: The perception of an image by a human observer is usually modeled as a parallel process in which all parts of the image are treated more or less equivalently, but in reality the analysis of scenes is a highly selective procedure, in which only a small subset of image locations is processed by the precise and efficient neural machinery of foveal vision. To understand the principles behind this selection of the ‘‘informative’’ regions of images, we have developed a hybrid system that consists of a combination of a knowledgebased reasoning system with a low-level preprocessing by linear and nonlinear neural operators. This hybrid system is intended as a first step towards a complete model of the sensorimotor system of saccadic scene analysis. In the analysis of a scene, the system calculates in each step which eye movement has to be made to reach a maximum of information about the scene. The possible information gain is calculated by means of a parallel strategy which is suitable for adaptive reasoning. The output of the system is a fixation sequence, and finally, a hypothesis about the scene.

Provably-convergent iterative methods for projective structure from motion

Abstract: The estimation of the projective structure of a scene from image correspondences can be formulated as the minimization of the mean-squared distance between predicted and observed image points with respect to the projection matrices, the scene point positions, and their depths. Since these unknowns are not independent, constraints must be chosen to ensure that the optimization process. is well posed. This paper examines three plausible choices, and shows that the first one leads to the Sturm-Triggs projective factorization algorithm, while the other two lead to new provably-convergent approaches. Experiments with synthetic and real data are used to compare the proposed techniques to the Sturm-Triggs algorithm and bundle adjustment.

Tracking nonrigid motion and structure from 2D satellite cloud images without correspondences

Abstract: Tracking both structure and motion of nonrigid objects from monocular images is an important problem in vision. In this paper, a hierarchical method which integrates local analysis (that recovers small details) and global analysis (that appropriately limits possible nonrigid behaviors) is developed to recover dense depth values and nonrigid motion from a sequence of 2D satellite cloud images without any prior knowledge of point correspondences. This problem is challenging not only due to the absence of correspondence information but also due to the lack of depth cues in the 2D cloud images (scaled orthographic projection). In our method, the cloud images are segmented into several small regions and local analysis is performed for each region. A recursive algorithm is proposed to integrate local analysis with appropriate global fluid model constraints, based on which a structure and motion analysis system, SMAS, is developed. We believe that this is the first reported system in estimating dense structure and nonrigid motion under scaled orthographic views using fluid model constraints. Experiments on cloud image sequences captured by meteorological satellites (GOES-8 and GOES-9) have been performed using our system, along with their validation and analyses. Both structure and 3D motion correspondences are estimated to subpixel accuracy. Our results are very encouraging and have many potential applications in earth and space sciences, especially in cloud models for weather prediction.

Multi-view scene capture by surfel sampling: from video streams to non-rigid 3D motion, shape and reflectance

Abstract: In this paper we study the problem of recovering the 3D shape, reflectance, and non-rigid motion of a dynamic 3D scene. Because these properties are completely unknown, our approach uses multiple views to build a piecewise continuous geometric and radiometric representation of the scene's trace in space-time. Basic primitive of this representation is the dynamic surfel, which (1) encodes the instantaneous local shape, reflectance, and motion of a small region in the scene, and (2) enables accurate prediction of the region's dynamic appearance under known illumination conditions. We show that complete surfel-based reconstructions can be created by repeatedly applying an algorithm called surfel sampling that combines sampling and parameter estimation to fit a single surfel to a small, bounded region of space-time. Experimental results with the Phong reflectance model and complex real scenes (clothing, skin, shiny objects) illustrate our method's ability to explain pixels and pixel variations in terms of their physical causes-shape, reflectance, motion, illumination, and visibility.

Time-Continuous Segmentation of Cardiac Image Sequences Using Active Appearance Motion Models

Abstract: This paper describes a novel, 2D+time Active Appearance Motion Model (AAMM). Cootes's 2D AAM framework was extended by considering a complete image sequence as a single shape/intensity sample. This way, the cardiac motion is modeled in combination with the shape and image appearance of the heart. The clinical potential of the AAMMs is demonstrated on two imaging modalities - cardiac MRI and echocardiography.

A tensor-driven active contour model for moving object segmentation

Abstract: We propose an approach to the segmentation of video objects based on motion cues. Motion analysis is performed by estimating local orientations in the spatiotemporal domain using the three-dimensional structure tensor. These estimates are integrated as an external force into an active contour model, thus stopping the evolving curve when it reaches the moving object's boundary. To enable simultaneous detection of several objects, we reformulate the tensor-based active contour model using the level-set technique. In addition, a contour refinement technique has been developed to better approximate the real boundary of the moving object. We provide promising experimental results calculated on real-world video sequences widely used within the computer vision community.

A Coordinate-Invariant Approach to Multiresolution Motion Analysis

Abstract: Multiresolution motion analysis has gained considerable research interest as a unified framework to facilitate a variety of motion editing tasks. Within this framework, motion data are represented as a collection of coefficients that form a coarse-to-fine hierarchy. The coefficients at the coarsest level describe the global pattern of a motion signal, while those at fine levels provide details at successively finer resolutions. Due to the inherent nonlinearity of the orientation space, the challenge is to generalize multiresolution representations for motion data that contain orientations as well as positions. Our goal is to develop a multiresolution analysis method that guarantees coordinate-invariance without singularity. To do so, we employ two novel ideas: hierarchical displacement mapping and motion filtering. Hierarchical displacement mapping provides an elegant formulation to describe positions and orientations in a coherent manner. Motion filtering enables us to separate motion details level-by-level to build a multiresolution representation in a coordinate-invariant way. Our representation facilitates multiresolution motion editing through level-wise coefficient manipulation that uniformly addresses issues raised by motion modification, blending, and stitching.

Method and apparatus for performing motion analysis on an image sequence

Abstract: A method and concomitant apparatus for performing motion analysis on a sequence of images is disclosed. Initially, the sequence of images is received from a video source. The sequence of images captures a plurality of objects each moving along a trajectory in an area imaged by the video source. Motion information is extracted from the sequence of images for each of the plurality of objects. Spatial patterns are then determined from the extracted motion information.

Moving object localisation using a multi-label fast marching algorithm

Abstract: In this paper, we address two problems crucial to motion analysis: the detection of moving objects and their localisation. Statistical and level set approaches are adopted in formulating these problems. For the change detection problem, the inter-frame difference is modelled by a mixture of two zero-mean Laplacian distributions. At first, statistical tests using criteria with negligible error probability are used for labelling as changed or unchanged as many sites as possible. All the connected components of the labelled sites are used thereafter as region seeds, which give the initial level sets for which velocity fields for label propagation are provided. We introduce a new multi-label fast marching algorithm for expanding competitive regions. The solution of the localisation problem is based on the map of changed pixels previously extracted. The boundary of the moving object is determined by a level set algorithm, which is initialised by two curves evolving in converging opposite directions. The sites of curve contact determine the position of the object boundary. Experimental results using real video sequences are presented, illustrating the efficiency of the proposed approach.

Systems and methods for determining eye glances

Abstract: A method determines driver glance information from an input video by performing motion analysis on the video; and performing image analysis on a frame of the video.

Key-posture extraction out of human motion data

Abstract: Recent progress in 3-D capture technology has made it possible to obtain much of realistic motion data of human subjects. Being captured in high frame rates, compression or extraction of key postures out of the motion data is useful for storage, transfer and browsing among them: this can serve as an Important pre-processing for applications such as rehabilitation, ergonomics and sports physiology. This paper addresses these problems by treating the motion date as trajectory curves In a high-dimensional space and doing a novel application of a curve simplification algorithm, typically used for planar curves, to human motion data.

Recognising human and animal movement by symmetry

Abstract: We show how the symmetry of motion can be extracted by using the generalised symmetry operator for analysing motion and for gait recognition. This operator, rather than relying on the borders of a shape or on general appearance, locates features by their symmetrical properties. This approach is reinforced by the view from psychology that human gait is a symmetrical pattern of motion, and by other works. We applied our new method to compare animal gait, and for recognition by gait. Results show that the symmetry properties of gait appear to be unique and can indeed be used for analysis and for recognition. We have so far achieved promising recognition rates of over 95%. Performance analysis also suggests that symmetry enjoys practical advantages such as relative immunity to noise with capability to handle occlusion and as such might prove suitable for applications like clip-database browsing.

A 3-D marker-free system for the analysis of movement disabilities - an application to the legs

Abstract: Describes an approach allowing the analysis of human motion in 3D space. The system that we developed is composed of three CCD (charge-coupled device) cameras that capture synchronized image sequences of a human body in motion without the use of markers. Characteristic points belonging to the boundaries of the body in motion are first extracted from the initial images. 2D superquadrics are then adjusted on these points by a fuzzy clustering process. After that, the position of a 3D model based on a set of articulated superquadrics, each of them describing a part of the human body, is reconstructed. An optical flow process allows the prediction of the position of the model from its position at a previous time, and gives initial values for the fuzzy classification. The results that we present more specifically concern the analysis of movement disabilities of a human leg during gait. They are improved by using articulation-based constraints. The methodology can be used in human motion analysis for clinical applications.

Towards a unified framework for tracking and analysis of human motion

Abstract: We propose a framework for detecting, tracking and analyzing non-rigid motion based on learned motion patterns. The framework features an appearance based approach to represent the spatial information and hidden Markov models (HMM) to encode the temporal dynamics of the time varying visual patterns. The low level spatial feature extraction is fused with the temporal analysis, providing a unified spatio-temporal approach to common detection, tracking and classification problems. This is a promising approach for many classes of human motion patterns. Visual tracking is achieved by extracting the most probable sequence of target locations from a video stream using a combination of random sampling and the forward procedure from HMM theory. The method allows us to perform a set of important tasks such as activity recognition, gait-analysis and keyframe extraction. The efficacy of the method is shown on both natural and synthetic test sequences.

Automated film rhythm extraction for scene analysis

Abstract: This paper examines film rhythm, an important expressive element in motion pictures, based on our ongoing study to exploit film grammar as a broad computational framework for the task of automated film and video understanding. Of the many, more or less elusive, narrative devices contributing to film rhythm, this paper discusses motion characteristics that form the basis of our analysis, and presents novel computational models for extracting rhythmic patterns induced through a perception of motion. In our rhythm model, motion behaviour is classified as being either nonexistent, fluid or staccato for a given shot. Shot neighbourhoods in movies are then grouped by proportional makeup of these motion behavioural classes to yield seven high-level rhythmic arrangements that prove to be adept at indicating likely scene content (e.g. dialogue or chase sequence) in our experiments. Underlying causes for this level of codification in our approach are postulated from film grammar, and are accompanied by detailed demonstration from real movies for the purposes of clarification.

Automatic measurement of intensity of motion activity of video segments

Abstract: We present a psycho-visual and analytical framework for automatic measurement of motion activity in view sequences. We construct a test-set of video segments by carefully selecting video segments form the MPEG-7 video test set. We construct a ground truth, based on subjective test with naive subjects. We find that the subjects agree reasonably on the motion activity of video segments, which makes the ground truth reliable. We present a set of automatically extractable, known and novel, descriptors of motion activity based on different hypotheses about subjective perception of motion activity. We show that all the descriptors perform well against the ground truth. We find that the MPEG-7 motion activity descriptor, based on variance of motion vector magnitudes, is one of the best in overall performance over the test set.

Computational modeling of non-Fourier motion: further evidence for a single luminance-based mechanism

Abstract: It is generally assumed that the perception of non-Fourier motion requires the operation of some nonlinearity before motion analysis. We apply a computational model of biological motion processing to a class of non-Fourier motion stimuli designed to investigate nonlinearity in human visual processing. The model correctly detects direction of motion in these non-Fourier stimuli without recourse to any preprocessing nonlinearity. This demonstrates that the non-Fourier motion in some non-Fourier stimuli is directly available to luminance-based motion mechanisms operating on measurements of local spatial and temporal gradients.

Motion recognition using spatio-temporal random walks in sequence of 2D motion-related measurements

Abstract: This paper describes an original approach for non parametric motion analysis in image sequences. It relies on a statistical modeling of distributions of local motion-related measurements, computed over image sequences, resulting from spatio-temporal random walks. It handles in a single probabilistic framework both spatial and temporal properties of motion content. The important feature of our method is to make feasible the exact computation of conditional likelihood functions. We have carried out motion recognition experiments over a large set of real image sequences comprising various motion types.

Segmenting Traffic Scenes from Grey Level and Motion Information

Abstract: This paper is concerned with an efficient estimation and segmentation of 2D motion from image sequences, with the focus on traffic monitoring applications. In order to reduce the computational load to achieve real-time implementation, the proposed approach makes use of simplifying assumptions that the camera is stationary, and that the projection of vehicles motion on the image plane can be approximated by translation. We show that satisfactory results can be achieved even under such apparently restrictive assumptions. The use of 2D motion analysis and the pre-segmentation stage significantly reduces the computational load, and the region-based motion estimator gives robustness to noise and changes in the illumination conditions.

Fusion of physically-based registration and deformation modeling for nonrigid motion analysis

Abstract: In our previous work, we used finite element models to determine nonrigid motion parameters and recover unknown local properties of objects given correspondence data recovered with snakes or other tracking models. In this paper, we present a novel multiscale approach to recovery of nonrigid motion from sequences of registered intensity and range images. The main idea of our approach is that a finite element (FEM) model incorporating material properties of the object can naturally handle both registration and deformation modeling using a single model-driving strategy. The method includes a multiscale iterative algorithm based on analysis of the undirected Hausdorff distance to recover correspondences. The method is evaluated with respect to speed and accuracy. Noise sensitivity issues are addressed. Advantages of the proposed approach are demonstrated using man-made elastic materials and human skin motion. Experiments with regular grid features are used for performance comparison with a conventional approach (separate snakes and FEM models). It is shown, however, that the new method does not require a sampling/correspondence template and can adapt the model to available object features. Usefulness of the method is presented not only in the context of tracking and motion analysis, but also for a burn scar detection application.

A fast method for calibrating video-based motion analysers using only a rigid bar

Abstract: Video-camera systems are widely used in biomechanics and clinical fields to measure the 3D kinematic measurements of human motion. To be used, they need to be calibrated, that is the parameters which geometrically define the cameras have to be determined. It is shown here how this can be achieved by surveying a rigid bar in motion inside the working volume, and in a very short time: less than 15 s on a Pentium III. The exterior parameters are estimated through the coplanarity constraint, the camera focal lengths through the properties of epipolar geometry and the principal points with a fast evolutionary optimisation which guarantees convergence when the initial principal points cannot be adequately estimated. The method has been widely tested on simulated and real data. Results show that its accuracy is comparable with that obtained using methods based on points of known 3D coordinates (DLT): 0.37 mm RMS error over a volume with a diagonal ≈1.5m. A preferential absolute reference system is obtained from the same bar motion data and is used to guide an intelligent decimation of the data. Finally, the role that the principal points play in achieving a high accuracy, which is questioned in the computer vision domain, is assessed through simulations.

Complex event classification in degraded image sequences

Abstract: We address the problem of motion estimation failure in degraded image sequences. This failure is caused by some complex events that take place in the image. As a result, the sequence of operations that rely on the motion estimation process, such as motion compensation and motion picture restoration, fail as well. The statistical analysis of the complex event areas indicates that it is possible to discriminate between the complex events resulting from complicated object motion and the ones resulting from image artefacts. An analysis scheme based on segment matching is proposed for the task of classifying the detected complex event areas.

A high-precision camera operation parameter measurement system and its application to image motion inferring

Abstract: Information about camera operations such as zoom, focus, pan, tilt and dollying is significant not only for efficient video coding, but also for content-based video representation. In this paper we describe a high-precision camera operation parameter measurement system and apply it to image motion inferring. First, we outline the implemented system which is designed to provide camera operation parameters with a high precision required for image coding applications. Second, we calibrate the camera lens to determine its exact optical properties, A pin-hole camera model with the 2nd order radial lens distortion and a two-image calibration technique are employed. Finally, we use the pan, tilt and zoom parameters measured by the system to infer image motion. The experimental results show that the inferred motion coincides with the actual motion very closely. Compared to the motion analysis techniques that estimate camera motion from video sequences, our approach does not suffer from ambiguity, thus can provide reliable and accurate image global motion. The obtained motion can be applied to image mosaicing, moving object segmentation, object-based image coding, etc.