Shawn Lankton

Localizing Region-Based Active Contours

In this paper, we propose a natural framework that allows any region-based segmentation energy to be re-formulated in a local way. We consider local rather than global image statistics and evolve a contour based on local information. Localized contours are capable of segmenting objects with heterogeneous feature profiles that would be difficult to capture correctly using a standard global method. The presented technique is versatile enough to be used with any global region-based active contour energy and instill in it the benefits of localization. We describe this framework and demonstrate the localization of three well-known energies in order to illustrate how our framework can be applied to any energy. We then compare each localized energy to its global counterpart to show the improvements that can be achieved. Next, an in-depth study of the behaviors of these energies in response to the degree of localization is given. Finally, we show results on challenging images to illustrate the robust and accurate segmentations that are possible with this new class of active contour models.

Hybrid geodesic region-based curve evolutions for image segmentation

In this paper we present a gradient descent flow based on a novel energy functional that is capable of producing robust and accurate segmentations of medical images. This flow is a hybridization of local geodesic active contours and more global region-based active contours. The combination of these two methods allows curves deforming under this energy to find only significant local minima and delineate object borders despite noise, poor edge information, and heterogeneous intensity profiles. To accomplish this, we construct a cost function that is evaluated along the evolving curve. In this cost, the value at each point on the curve is based on the analysis of interior and exterior means in a local neighborhood around that point. We also demonstrate a novel mathematical derivation used to implement this and other similar flows. Results for this algorithm are compared to standard techniques using medical and synthetic images to demonstrate the proposed methods robustness and accuracy as compared to both edge-based and region-based alone.

Localized statistics for DW-MRI fiber bundle segmentation

We describe a method for segmenting neural fiber bundles in diffusion-weighted magnetic resonance images (DW-MRI). As these bundles traverse the brain to connect regions, their local orientation of diffusion changes drastically, hence a constant global model is inaccurate. We propose a method to compute localized statistics on orientation information and use it to drive a variational active contour segmentation that accurately models the non-homogeneous orientation information present along the bundle. Initialized from a single fiber path, the proposed method proceeds to capture the entire bundle. We demonstrate results using the technique to segment the cingulum bundle and describe several extensions making the technique applicable to a wide range of tissues.

Object Tracking and Target Reacquisition Based on 3-D Range Data for Moving Vehicles

In this paper, we propose an approach for tracking an object of interest based on 3-D range data. We employ particle filtering and active contours to simultaneously estimate the global motion of the object and its local deformations. The proposed algorithm takes advantage of range information to deal with the challenging (but common) situation in which the tracked object disappears from the image domain entirely and reappears later. To cope with this problem, a method based on principle component analysis (PCA) of shape information is proposed. In the proposed method, if the target disappears out of frame, shape similarity energy is used to detect target candidates that match a template shape learned online from previously observed frames. Thus, we require no a priori knowledge of the targets shape. Experimental results show the practical applicability and robustness of the proposed algorithm in realistic tracking scenarios.

Fast Optimal Mass Transport for Dynamic Active Contour Tracking on the GPU

In computational vision, visual tracking remains one of the most challenging problems due to noise, clutter, occlusion, and dynamic scenes. No one technique has yet managed to solve this problem completely, but those that employ control- theoretic filtering techniques have proven to be quite successful. In this work, we extend one such technique by Niethammer et al. in which implicitly represented dynamically evolving contours are filtered using a geometric observer framework. The effectiveness of the observer hangs upon the solution of two major problems: (1) the calculation of accurate curve velocities and (2) the determination of diffeomorphic correspondence maps between curves for geometric interpolation. We propose the use of novel image registration techniques such as image warping and optimal mass transport for the solution of these problems which increase the performance of the framework and reduce algorithmic complexity. One major drawback to the original scheme, as it relies on PDE solutions, is its computational burden restricting it from real time use. We show that the framework can, in fact, run in near real time by implementing our additions to the framework on the graphics processing unit (GPU) and show better execution times for these algorithms than reported in recent literature.

Tracking through changes in scale

We propose a tracking system that is especially well-suited to tracking targets which change drastically in size or appearance. To accomplish this, we employ a fast, two phase template matching algorithm along with a periodic template update method. The template matching step ensures accurate localization while the template update scheme allows the target model to change over time along with the appearance of the target. Furthermore, the algorithm can deliver real-time results even when targets are very large. We demonstrate the proposed method with good results on several sequences showing targets which exhibit large changes in size, shape, and appearance.

TAC: Thresholding active contours

In this paper, we describe a region-based active contour technique to perform image segmentation. We propose an energy functional that realizes an explicit trade-off between the (current) image segmentation obtained from a curve and the (implied) segmentation obtained from dynamically thresholding the image. In contrast with standard region-based techniques, the resulting variational approach bypasses the need to fit (a priori chosen) statistical models to the object and the background. Our technique performs segmentation based on geometric considerations of the image and contour, instead of statistical ones. The resulting flow leads to very reasonable segmentations as shown by several illustrative examples.

Improved tracking by decoupling camera and target motion

Video tracking is widely used for surveillance, security, and defense purposes. In cases where the camera is not fixed due to pans and tilts, or due to being fixed on a moving platform, tracking can become more difficult. Camera motion must be taken into account, and objects that come and go from the field of view should be continuously and uniquely tracked. We propose a tracking system that can meet these needs by using a frame registration technique to estimate camera motion. This estimate is then used as the input control signal to a Kalman filter which estimates the targets motion model based on measurements from a mean-shift localization scheme. Thus we decouple the camera and object motion and recast the problem in terms of a principled control theory solution. Our experiments show that using a controller built on these principles we are able to track videos with multiple objects in sequences with moving cameras. Furthermore, the techniques are computationally efficient and allow us to accomplish these results in real-time. Of specific importance is that when objects are lost off-frame they can still be uniquely identified and reacquired when they return to the field of view.

A robust aim point tracking algorithm for 3-D laser radar imagery

In this work, we present a controlled active vision tracking scheme that makes use of 3-D range data and 2-D reflectance data collected with a 3-D Laser Radar (3DLADAR) system. Specifically, our algorithm employs the Active Contour Framework along with a detection algorithm involving feedback to effectively maintain visual tracking of an object in adverse scenarios. As opposed to typical 2D systems, in which tracking is limited by a lack of contrast, 3DLADAR provides the capability to improve aim point tracking by encoding depth information so that the scene can be resolved for all spatial dimensions. We demonstrate the proposed algorithm both qualitatively and quantitatively on several challenging tracking scenarios.

Statistical shape learning for 3D tracking

In this note, we consider the use of 3D models for visual tracking in controlled active vision. The models are used for a joint 2D segmentation/3D pose estimation procedure in which we automatically couple the two processes under one energy functional. Further, employing principal component analysis from statistical learning, can train our tracker on a catalog of 3D shapes, giving a priori shape information. The segmentation itself is information-based. The allows us to track in uncertain adversarial environments. Our methodology is demonstrated on some real sequences which illustrate its robustness on challenging scenarios.