Ruzena Bajcsy

Multiresolution elastic matching

Matching of locally variant data to an explicit 3-dimensional pictorial model is developed for X-ray computed tomography scans of the human brain, where the model is a voxel representation of an anatomical human brain atlas. The matching process is 3-dimensional without any preference given to the slicing plane. After global alignment the brain atlas is deformed like a piece of rubber, without tearing or folding. Deformation proceeds step-by-step in a coarse-to-fine strategy, increasing the local similarity and global coherence. The assumption underlying this approach is that all normal brains, at least at a certain level of representation, have the same topological structure, but may differ in shape details. Results show that we can account for these differences.

Recovery of parametric models from range images: the case for superquadrics with global deformations

A method for recovery of compact volumetric models for shape representation of single-part objects in computer vision is introduced. The models are superquadrics with parametric deformations (bending, tapering, and cavity deformation). The input for the model recovery is three-dimensional range points. Model recovery is formulated as a least-squares minimization of a cost function for all range points belonging to a single part. During an iterative gradient descent minimization process, all model parameters are adjusted simultaneously, recovery position, orientation, size, and shape of the model, such that most of the given range points lie close to the models surface. A specific solution among several acceptable solutions, where are all minima in the parameter space, can be reached by constraining the search to a part of the parameter space. The many shallow local minima in the parameter space are avoided as a solution by using a stochastic technique during minimization. Results using real range data show that the recovered models are stable and that the recovery procedure is fast. >

Congestion control and fairness for many-to-one routing in sensor networks

In this paper we propose a distributed and scalable algorithm that eliminates congestion within a sensor network, and that ensures the fair delivery of packets to a central node, or base station. We say that fairness is achieved when equal number of packets are received from each node. Since in general we have many sensors transmitting data to the base station, we consider the scenario where we have many-to-one multihop routing, noting that it can easily be extended to unicast or many-to-many routing. Such routing structures often result in the sensors closer to the base station experiencing congestion, which inevitably cause packets originating from sensors further away from the base station to have a higher probability of being dropped. Our algorithm exists in the transport layer of the traditional network stack model, and is designed to work with any MAC protocol in the data-link layer with minor modifications. Our solution is scalable, each sensor mote requires state proportional to the number of its neighbors. Finally, we demonstrate the effectiveness of our solution with both simulations and actual implementation in UC Berkeleys sensor motes.

Elastically deforming 3D atlas to match anatomical brain images

To evaluate our system for elastically deforming a three-dimensional atlas to match anatomical brain images, six deformed versions of an atlas were generated. The deformed atlases were created by elastically mapping an anatomical brain atlas onto different MR brain image volumes. The mapping matches the edges of the ventricles and the surface of the brain; the resultant deformations are propagated through the atlas volume, deforming the remainder of the structures in the process. The atlas was then elastically matched to its deformed versions. The accuracy of the resultant matches was evaluated by determining the correspondence of 32 cortical and subcortical structures. The system on average matched the centroid of a structure to within 1 mm of its true position and fit a structure to within 11% of its true volume. The overlap between the matched and true structures, defined by the ratio between the volume of their intersection and the volume of their union, averaged 66%. When the gray-white interface was included for matching, the mean overlap improved to 78%; each structure was matched to within 0.6 mm of its true position and fit to within 6% of its true volume. Preliminary studies were also made to determine the effect of the compliance of the atlas on the resultant match.

Wearable Sensors for Reliable Fall Detection

Unintentional falls are a common cause of severe injury in the elderly population. By introducing small, non-invasive sensor motes in conjunction with a wireless network, the Ivy Project aims to provide a path towards more independent living for the elderly. Using a small device worn on the waist and a network of fixed motes in the home environment, we can detect the occurrence of a fall and the location of the victim. Low-cost and low-power MEMS accelerometers are used to detect the fall while RF signal strength is used to locate the person

Occlusions as a guide for planning the next view

A strategy for acquiring 3-D data of an unknown scene, using range images obtained by a light stripe range finder is addressed. The foci of attention are occluded regions, i.e., only the scene at the borders of the occlusions is modeled to compute the next move. Since the system has knowledge of the sensor geometry, it can resolve the appearance of occlusions by analyzing them. The problem of 3-D data acquisition is divided into two subproblems due to two types of occlusions. An occlusion arises either when the reflected laser light does not reach the camera or when the directed laser light does not reach the scene surface. After taking the range image of a scene, the regions of no data due to the first kind of occlusion are extracted. The missing data are acquired by rotating the sensor system in the scanning plane, which is defined by the first scan. After a complete image of the surface illuminated from the first scanning plane has been built, the regions of missing data due to the second kind of occlusions are located. Then, the directions of the next scanning planes for further 3-D data acquisition are computed. >

Texture gradient as a depth cue

A texture operator for use in computer vision programs is described. The operator classifies texture according to characteristics of the Fourier transform of local image windows. Gradients of the texture are found by comparing and associating quantitative and qualitative values of adjacent windows. The gradients are then interpreted as a depth cue for longitudinal (receding) surfaces. Experimental results with natural, outdoor scenes are reported.

A Computerized System for the Elastic Matching of Deformed Radiographic Images to Idealized Atlas Images

A system of computer programs is described which, for the first time, is able to use computerized tomographic data to automatically locate, measure, and describe anatomical structures of interest with accuracy and consistency. Input to the system consists of any digitized radiographic data. Computer assisted tomographic (CAT) scans of the head were used in this first implementation. Using these data and a predefined atlas picture representing an idealized view of the average normal image, an individualized atlas was created. From the individualized atlas, structure size, density, location displacement, and distortion may be calculated. The individualized atlas created using high resolution data, such as the CAT scan, may then be directly superimposed on pictures obtained using lower resolution modalities, such as positron emission tomographic scan images. This allows the precise location of structures poorly visualized by the secondary imaging modality. This system is capable of using either two- or three-dimensional data.

Accuracy and robustness of Kinect pose estimation in the context of coaching of elderly population

The Microsoft Kinect camera is becoming increasingly popular in many areas aside from entertainment, including human activity monitoring and rehabilitation. Many people, however, fail to consider the reliability and accuracy of the Kinect human pose estimation when they depend on it as a measuring system. In this paper we compare the Kinect pose estimation (skeletonization) with more established techniques for pose estimation from motion capture data, examining the accuracy of joint localization and robustness of pose estimation with respect to the orientation and occlusions. We have evaluated six physical exercises aimed at coaching of elderly population. Experimental results present pose estimation accuracy rates and corresponding error bounds for the Kinect system.

Segmentation of range images as the search for geometric parametric models

Segmentation of range images has long been considered in computer vision as an important but extremely difficult problem. In this paper we present a new paradigm for the segmentation of range images into piecewise continuous surfaces. Data aggregation is performed via model recovery in terms of variable-order bi-variate polynomials using iterative regression. Model recovery is initiated independently in regularly placed seed regions in the image. All the recovered models are potential candidates for the final description of the data. Selection of the models is defined as a quadratic Boolean problem, and the solution is sought by the WTA (winner-takes-all) technique, which turns out to be a good compromise between the speed of computation and the accuracy of the solution. The overall efficiency of the method is achieved by combining model recovery and model selection in an iterative way. Partial recovery of the models is followed by the selection (optimization) procedure and only the “best” models are allowed to develop further.The major novelty of the approach lies in an effective combination of simple component algorithms, which stands in contrast to methods which attempt to solve the problem in a single processing step using sophisticated means. We present the results on several real range images.