Improved Tracking of Multiple Humans with Trajectory Prediction and Occlusion Modeling

A combined 2D, 3D approach is presented that allows for robust tracking of moving bodies in a given environment as observed via a single, uncalibrated video camera. Tracking is robust even in the presence of occlusions. Low-level features are often insufficient for detection, segmentation, and tracking of non-rigid moving objects. Therefore, an improved mechanism is proposed that combines low-level (image processing) and mid-level (recursive trajectory estimation) information obtained during the tracking process. The resulting system can segment and maintain the tracking of moving objects before, during, and after occlusion. At each frame, the system also extracts a stabilized coordinate frame of the moving objects. This stabilized frame is used to resize and resample the moving blob so that it can be used as input to motion recognition modules. The approach enables robust tracking without constraining the system to know the shape of the objects being tracked beforehand; although, some assumptions are made about the characteristics of the shape of the objects, and how they evolve with time. Experiments in tracking moving people are described.

3D Trajectory Recovery for Tracking Multiple Objects and Trajectory Guided Recognition of Actions

A mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. The 3D trajectory, occlusion, and segmentation information are utilized in extracting stabilized views of the moving object. Trajectory-guided recognition (TGR) is proposed as a new and efficient method for adaptive classification of action. The TGR approach is demonstrated using "motion history images" that are then recognized via a mixture of Gaussian classifier. The system was tested in recognizing various dynamic human outdoor activities; e.g., running, walking, roller blading, and cycling. Experiments with synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise.

Trajectory Guided Tracking and Recognition of Actions

A combined 2D, 3D approach is presented that allows for robust tracking of moving people and recognition of actions. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. Low-level features are often insufficient for detection, segmentation, and tracking of non-rigid moving objects. Therefore, an improved mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. The 3D trajectory, occlusion, and segmentation information are utilized in extracting stabilized views of the moving object that are then used as input to action recognition modules. Trajectory-guided recognition (TGR) is proposed as a new and efficient method for adaptive classification of action. The TGR approach is demonstrated using "motion history images" that are then recognized via a mixture-of-Gaussians classifier. The system was tested in recognizing various dynamic human outdoor activities: running, walking, roller blading, and cycling. Experiments with real and synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise.

Robust Tracking of Human Motion

This technical report presents a combined solution for two problems, one: tracking objects in 3D space and estimating their trajectories and second: computing the similarity between previously estimated trajectories and clustering them using the similarities that we just computed. For the first part, trajectories are estimated using an EKF formulation that will provide the 3D trajectory up to a constant. To improve accuracy, when occlusions appear, multiple hypotheses are followed. For the second problem we compute the distances between trajectories using a similarity based on LCSS formulation. Similarities are computed between projections of trajectories on coordinate axes. Finally we group trajectories together based on previously computed distances, using a clustering algorithm. To check the validity of our approach, several experiments using real data were performed.

Extraction and Clustering of Motion Trajectories in Video

A system is described that tracks moving objects in a video dataset so as to extract a representation of the objects' 3D trajectories. The system then finds hierarchical clusters of similar trajectories in the video dataset. Objects' motion trajectories are extracted via an EKF formulation that provides each object's 3D trajectory up to a constant factor. To increase accuracy when occlusions occur, multiple tracking hypotheses are followed. For trajectory-based clustering and retrieval, a modified version of edit distance, called longest common subsequence (LCSS) is employed. Similarities are computed between projections of trajectories on coordinate axes. Trajectories are grouped based, using an agglomerative clustering algorithm. To check the validity of the approach, experiments using real data were performed.

An Invariant Representation for Matching Trajectories across Uncalibrated Video Streams

We introduce a view-point invariant representation of moving object trajectories that can be used in video database applications. It is assumed that trajectories lie on a surface that can be locally approximated with a plane. Raw trajectory data is first locally approximated with a cubic spline via least squares fitting. For each sampled point of the obtained curve, a projective invariant feature is computed using a small number of points in its neighborhood. The resulting sequence of invariant features computed along the entire trajectory forms the view invariant descriptor of the trajectory itself. Time parametrization has been exploited to compute cross ratios without ambiguity due to point ordering. Similarity between descriptors of different trajectories is measured with a distance that takes into account the statistical properties of the cross ratio, and its symmetry with respect to the point at infinity. In experiments, an overall correct classification rate of about 95% has been obtained on a dataset of 58 trajectories of players in soccer video, and an overall correct classification rate of about 80% has been obtained on matching partial segments of trajectories collected from two overlapping views of outdoor scenes with moving people and cars.

View Registration Using Interesting Segments of Planar Trajectories

We introduce a method for recovering the spatial and temporal alignment between two or more views of objects moving over a ground plane. Existing approaches either assume that the streams are globally synchronized, so that only solving the spatial alignment is needed, or that the temporal misalignment is small enough so that exhaustive search can be performed. In contrast, our approach can recover both the spatial and temporal alignment. We compute for each trajectory a number of interesting segments, and we use their description to form putative matches between trajectories. Each pair of corresponding interesting segments induces a temporal alignment, and defines an interval of common support across two views of an object that is used to recover the spatial alignment. Interesting segments and their descriptors are defined using algebraic projective invariants measured along the trajectories. Similarity between interesting segments is computed taking into account the statistics of such invariants. Candidate alignment parameters are verified checking the consistency, in terms of the symmetric transfer error, of all the putative pairs of corresponding interesting segments. Experiments are conducted with two different sets of data, one with two views of an outdoor scene featuring moving people and cars, and one with four views of a laboratory sequence featuring moving radio-controlled cars.

Neural Models of Motion Integration, Segmentation, and Probablistic Decision-Making

When brain mechanism carry out motion integration and segmentation processes that compute unambiguous global motion percepts from ambiguous local motion signals? Consider, for example, a deer running at variable speeds behind forest cover. The forest cover is an occluder that creates apertures through which fragments of the deer's motion signals are intermittently experienced. The brain coherently groups these fragments into a trackable percept of the deer in its trajectory. Form and motion processes are needed to accomplish this using feedforward and feedback interactions both within and across cortical processing streams. All the cortical areas V1, V2, MT, and MST are involved in these interactions. Figure-ground processes in the form stream through V2, such as the seperation of occluding boundaries of the forest cover from the boundaries of the deer, select the motion signals which determine global object motion percepts in the motion stream through MT. Sparse, but unambiguous, feauture tracking signals are amplified before they propogate across position and are intergrated with far more numerous ambiguous motion signals. Figure-ground and integration processes together determine the global percept. A neural model predicts the processing stages that embody these form and motion interactions. Model concepts and data are summarized about motion grouping across apertures in response to a wide variety of displays, and probabilistic decision making in parietal cortex in response to random dot displays.

An Active Pattern Recognition Architecture for Mobile Robots

An active, attentionally-modulated recognition architecture is proposed for object recognition and scene analysis. The proposed architecture forms part of navigation and trajectory planning modules for mobile robots. Key characteristics of the system include movement planning and execution based on environmental factors and internal goal definitions. Real-time implementation of the system is based on space-variant representation of the visual field, as well as an optimal visual processing scheme utilizing separate and parallel channels for the extraction of boundaries and stimulus qualities. A spatial and temporal grouping module (VWM) allows for scene scanning, multi-object segmentation, and featural/object priming. VWM is used to modulate a tn~ectory formation module capable of redirecting the focus of spatial attention. Finally, an object recognition module based on adaptive resonance theory is interfaced through VWM to the visual processing module. The system is capable of using information from different modalities to disambiguate sensory input.

The VITEWRITE Model of Handwriting Production

This article describes the VITEWRITE model for generating handwriting movements. The model consists of a sequential controller, or motor program, that interacts with a trajectory generator to move a hand with redundant degrees of freedom. The neural trajectory generator is the Vector Integration to Endpoint (VITE) model for synchronous variable-speed control of multijoint movements. VITE properties enable a simple control strategy to generate complex handwritten script if the hand model contains redundant degrees of freedom. The controller launches transient directional commands to independent hand synergies at times when the hand begins to move, or when a velocity peak in the outflow command to a given synergy occurs. The VITE model translates these temporally disjoint synergy commands into smooth curvilinear trajectories among temporally overlapping synergetic movements. Each synergy exhibits a unimodal velocity profile during any stroke, generates letters that are invariant under speed and size rescaling, and enables effortless connection of letter shapes into words. Speed and size rescaling are achieved by scalar GO and GRO signals that express computationally simple volitional commands. Psychophysical data such as the isochrony principle, asymmetric velocity profiles, and the two-thirds power law relating movement curvature and velocity arise as emergent properties of model interactions.

Navite: A Neural Network System For Sensory-Based Robot Navigation

A neural network system, NAVITE, for incremental trajectory generation and obstacle avoidance is presented. Unlike other approaches, the system is effective in unstructured environments. Multimodal inforrnation from visual and range data is used for obstacle detection and to eliminate uncertainty in the measurements. Optimal paths are computed without explicitly optimizing cost functions, therefore reducing computational expenses. Simulations of a planar mobile robot (including the dynamic characteristics of the plant) in obstacle-free and object avoidance trajectories are presented. The system can be extended to incorporate global map information into the local decision-making process.

Unsupervised Neural Network for the Control of a Mobile Robot

This article introduces an unsupervised neural architecture for the control of a mobile robot. The system allows incremental learning of the plant during robot operation, with robust performance despite unexpected changes of robot parameters such as wheel radius and inter-wheel distance. The model combines Vector associative Map (VAM) learning and associate learning, enabling the robot to reach targets at arbitrary distances without knowledge of the robot kinematics and without trajectory recording, but relating wheel velocities with robot movements.

Vector Associative Maps: Unsupervised Real-time Error-based Learning and Control of Movement Trajectories

This article describes neural network models for adaptive control of arm movement trajectories during visually guided reaching and, more generally, a framework for unsupervised real-time error-based learning. The models clarify how a child, or untrained robot, can learn to reach for objects that it sees. Piaget has provided basic insights with his concept of a circular reaction: As an infant makes internally generated movements of its hand, the eyes automatically follow this motion. A transformation is learned between the visual representation of hand position and the motor representation of hand position. Learning of this transformation eventually enables the child to accurately reach for visually detected targets. Grossberg and Kuperstein have shown how the eye movement system can use visual error signals to correct movement parameters via cerebellar learning. Here it is shown how endogenously generated arm movements lead to adaptive tuning of arm control parameters. These movements also activate the target position representations that are used to learn the visuo-motor transformation that controls visually guided reaching. The AVITE model presented here is an adaptive neural circuit based on the Vector Integration to Endpoint (VITE) model for arm and speech trajectory generation of Bullock and Grossberg. In the VITE model, a Target Position Command (TPC) represents the location of the desired target. The Present Position Command (PPC) encodes the present hand-arm configuration. The Difference Vector (DV) population continuously.computes the difference between the PPC and the TPC. A speed-controlling GO signal multiplies DV output. The PPC integrates the (DV)·(GO) product and generates an outflow command to the arm. Integration at the PPC continues at a rate dependent on GO signal size until the DV reaches zero, at which time the PPC equals the TPC. The AVITE model explains how self-consistent TPC and PPC coordinates are autonomously generated and learned. Learning of AVITE parameters is regulated by activation of a self-regulating Endogenous Random Generator (ERG) of training vectors. Each vector is integrated at the PPC, giving rise to a movement command. The generation of each vector induces a complementary postural phase during which ERG output stops and learning occurs. Then a new vector is generated and the cycle is repeated. This cyclic, biphasic behavior is controlled by a specialized gated dipole circuit. ERG output autonomously stops in such a way that, across trials, a broad sample of workspace target positions is generated. When the ERG shuts off, a modulator gate opens, copying the PPC into the TPC. Learning of a transformation from TPC to PPC occurs using the DV as an error signal that is zeroed due to learning. This learning scheme is called a Vector Associative Map, or VAM. The VAM model is a general-purpose device for autonomous real-time error-based learning and performance of associative maps. The DV stage serves the dual function of reading out new TPCs during performance and reading in new adaptive weights during learning, without a disruption of real-time operation. YAMs thus provide an on-line unsupervised alternative to the off-line properties of supervised error-correction learning algorithms. YAMs and VAM cascades for learning motor-to-motor and spatial-to-motor maps are described. YAM models and Adaptive Resonance Theory (ART) models exhibit complementary matching, learning, and performance properties that together provide a foundation for designing a total sensory-cognitive and cognitive-motor autonomous system.

Adaptive Neural Networks for Control of Movement Trajectories Invariant under Speed and Force Rescaling

This article describes two neural network modules that form part of an emerging theory of how adaptive control of goal-directed sensory-motor skills is achieved by humans and other animals. The Vector-Integration-To-Endpoint (VITE) model suggests how synchronous multi-joint trajectories are generated and performed at variable speeds. The Factorization-of-LEngth-and-TEnsion (FLETE) model suggests how outflow movement commands from a VITE model may be performed at variable force levels without a loss of positional accuracy. The invariance of positional control under speed and force rescaling sheds new light upon a familiar strategy of motor skill development: Skill learning begins with performance at low speed and low limb compliance and proceeds to higher speeds and compliances. The VITE model helps to explain many neural and behavioral data about trajectory formation, including data about neural coding within the posterior parietal cortex, motor cortex, and globus pallidus, and behavioral properties such as Woodworth's Law, Fitts Law, peak acceleration as a function of movement amplitude and duration, isotonic arm movement properties before and after arm-deafferentation, central error correction properties of isometric contractions, motor priming without overt action, velocity amplification during target switching, velocity profile invariance across different movement distances, changes in velocity profile asymmetry across different movement durations, staggered onset times for controlling linear trajectories with synchronous offset times, changes in the ratio of maximum to average velocity during discrete versus serial movements, and shared properties of arm and speech articulator movements. The FLETE model provides new insights into how spina-muscular circuits process variable forces without a loss of positional control. These results explicate the size principle of motor neuron recruitment, descending co-contractive compliance signals, Renshaw cells, Ia interneurons, fast automatic reactive control by ascending feedback from muscle spindles, slow adaptive predictive control via cerebellar learning using muscle spindle error signals to train adaptive movement gains, fractured somatotopy in the opponent organization of cerebellar learning, adaptive compensation for variable moment-arms, and force feedback from Golgi tendon organs. More generally, the models provide a computational rationale for the use of nonspecific control signals in volitional control, or "acts of will", and of efference copies and opponent processing in both reactive and adaptive motor control tasks.

Emergence of Tri-Phasic Muscle Activation from the Non-linear Interactions of Central and Spinal Neural Network Circuits

The origin of the tri-phasic burst pattern, observed in the EMGs of opponent muscles during rapid self-terminated movements, has been controversial. Here we show by computer simulation that the pattern emerges from interactions between a central neural trajectory controller (VITE circuit) and a peripheral neuromuscularforce controller (FLETE circuit). Both neural models have been derived from simple functional constraints that have led to principled explanations of a wide variety of behavioral and neurobiological data, including, as shown here, the generation of tri-phasic bursts.