Model-based Tracking Research Articles

Real-time tracking of free-form objects by range and intensity image fusion

Range images are helpful in tracking the motion of objects defined by smooth curved surfaces. However, combined use of range images with intensity data such as texture data is desirable in order to improve measurement stability. In this paper, a method is proposed for model-based tracking of 6-axis object motion. Pose estimation based on intensity image is not merely added to a range-image-based tracking system but instead, a texture-mapped model is built through range and intensity measurement and then used for tracking. We first explain a method for direct acquisition of object motion by applying the object rigidity to the gradient method, and then this method is shown to be easily fused with range-image-based motion tracking. An algorithm is also offered for high-speed image generation and reflectance estimation using CG hardware. Finally, experiments are described that demonstrate the feasibility of real-time tracking of the 6-axis motion of an object and of acquisition of the shape and reflectance model. © 1998 Scripta Technica, Syst Comp Jpn, 29(8): 19–27, 1998

Tracking of complex driving manoeuvres in traffic image sequences

It has been shown in previous publications (Proceedings of the Fourth European Conference on Computer Vision 1996 (ECCV '96), 14–18 April 1996, Cambridge, UK, Lecture Notes in Computer Science 1065 (Vol. II), Springer-Verlag, Berlin, Heidelberg, 1996, pp. 388–399, pp. 485–494) that model-based tracking of partially occluded vehicles in image sequences can be significantly improved by incorporating optical flow measurements on the one hand and by explicitly modelling occlusions on the other hand. In this contribution, we slightly extend the first approach and combine it with the second in order to demonstrate that consequent exploitation of model knowledge will lead to robust tracking results even for complex driving manoeuvres, significant occlusions, low contrast and small vehicle images.

Model-based tracking of complex innercity road intersections

Vision-based automatic driving along innercity roads and across complex innercity intersections requires to detect and track road markings and lane boundaries in order to determine the position and orientation of the vehicle relative to the ground. The complexity of intersection scenes and the disturbances in the detected contours enforce the use of model-based state estimation techniques. We recorded monocular image sequences of complex innercity road intersections from a moving small experimental truck in order to track intersection models using a Kalman filter. A scalar distance measure for the distance between image contour points and model edge segments turned out to be advantageous for the estimation process. This scalar measure is based on the simultaneous exploitation of edge element location and direction, thus using more information about the image gray value variation than previously when only the perpendicular distance between the edge element and the model segment was taken into account. We report about the theoretical foundation of this approach, its implementation and experimental results from several intersection sequences, as well as detailed comparisons with a previous, less sophisticated approach. These examples demonstrate the kind of problems which are likely to occur more frequently at intersection scenes than at those road scenes encountered while following a more or less straight, uninterrupted highway lane with clearly marked boundaries.

Sign language recognition using model-based tracking and a 3D Hopfield neural network

This paper presents a sign language recognition system which consists of three modules: model-based hand tracking, feature extraction, and gesture recognition using a 3D Hopfield neural network (HNN). The first one uses the Hausdorff distance measure to track shape-variant hand motion, the second one applies the scale and rotation-invariant Fourier descriptor to characterize hand figures, and the last one performs a graph matching between the input gesture model and the stored models by using a 3D modified HNN to recognize the gesture. Our system tests 15 different hand gestures. The experimental results show that our system can achieve above 91% recognition rate, and the recognition process time is about 10 s. The major contribution in this paper is that we propose a 3D modified HNN for gesture recognition which is more reliable than the conventional methods.

Model-based tracking of moving object

This paper describes a real-time tracking system which detects an object entering the field of view of a camera and executes tracking of the detected object by controlling a servo device in such a way that a target always lies at the center of an image frame. In order to detect and track a moving object, we basically apply model matching strategy. We allow the model of a target to vary dynamically during the tracking process so that it can assimilate variations of shape and intensities of a target. We also utilize a Kalman filter to encode a tracking history into state parameters of the filter. The estimated state parameters will then be used to reduce search areas for model matching and to control a servo device. Experimental results show that model adaptation allows robust tracking of a target object in dynamic environments. These experiments also confirm that the predicted values of a Kalman filter are very accurate in controlling a servo device and finding out the search areas for model matching. This paper concludes with theoretical bounds within which a tracking system can follow the movement of a target object.

Influence of an Explicitly Modelled 3D Scene on the Tracking of Partially Occluded Vehicles

Model-based tracking of vehicles in real world image sequences of traffic may fail due to different reasons. A careful analysis of failed tracking experiments brought to light that one of these phenomena consists in an incorrect match of parts of the vehicle model to image features belonging to other scene components. This effect appears in particular if the vehicle is occluded by either stationary scene components or by another moving vehicle. Although we got some encouraging results when we modeled the occluding scene components explicitly, we still encountered cases in which we did not succeed in tracking partially occluded vehicles properly. In this contribution we show some successful tracking results obtained from real world image sequences and discuss the cases of failures. Additionally, we present a framework for the conceptual characterization of occurring occlusions which became amenable to experimental analysis in the course of our investigations.

Learning the distribution of object trajectories for event recognition

The advent in recent years of robust, real-time, model-based tracking techniques for rigid and non-rigid moving objects has made automated surveillance and event recognition a possibility. A statistically based model of object trajectories is presented which is learnt from the observation of long image sequences. Trajectory data is supplied by a tracker using Active Shape Models, from which a model of the distribution of typical trajectories is learnt. Experimental results are included to show the generation of the model for trajectories within a pedestrian scene. We indicate how the resulting model can be used for the identification of atypical events.

Renewal models for maneuvering targets

Most model-based tracking algorithms represent the temporal dynamics of a maneuver acceleration with a Markov process. The sample paths so generated may not be plausible. A renewal process model is shown to be more realistic, but a tracker based upon it is seen to be more complicated to implement. The response of the renewal-model tracker is compared with a simpler estimator based upon a Markov model both for the case in which the tracker utilizes an imager, and the case where it does not. It is shown in the latter case that performance improvement is not commensurate with algorithmic complexity. For an image-enhanced tracker, a sophisticated model of target dynamics promises significant performance improvement. >

A new model-based tracking controller for robot manipulators using trajectory pattern inverse dynamics

A novel inverse-dynamic model-based tracking controller is presented. The inverse dynamics models are derived using the trajectory pattern method. The path of motion is divided into a number of relatively large segments, along which the motion is described by a selected trajectory pattern. The resulting inverse dynamics models are in the form of a summation of a fundamental sinusoidal time function and a number of its harmonics with constant coefficients. The coefficients are computed offline using analytical expressions that are derived for the selected trajectory pattern. The controller permits ultra-high-speed generation of the desired trajectory and feedforward signals. The online computations are minimal and can all be performed in parallel. The inverse-dynamic model of a spatial 3R manipulator is derived for a specified trajectory pattern. The controller structure and simulation results indicating its effectiveness are presented.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Real-time model-based tracking combining spatial and temporal features

This paper describes a method for tracking moving image features by combining spatial and temporal edge information with model-based feature information. The algorithm updates the two-dimensional position of object features by correlating predicted model features with current image data. The results of the correlation process are used to compute an updated model. The algorithm makes use of a high temporal sampling rate with respect to spatial changes of the image features and operates in a real-time multi-processing environment. Preliminary results demonstrate successful tracking for image feature velocities between 1.1 and 4.5 pixels every image frame.