Deformable Shape Detection and Description via Model-Based Region Grouping

A method for deformable shape detection and recognition is described. Deformable shape templates are used to partition the image into a globally consistent interpretation, determined in part by the minimum description length principle. Statistical shape models enforce the prior probabilities on global, parametric deformations for each object class. Once trained, the system autonomously segments deformed shapes from the background, while not merging them with adjacent objects or shadows. The formulation can be used to group image regions based on any image homogeneity predicate; e.g., texture, color, or motion. The recovered shape models can be used directly in object recognition. Experiments with color imagery are reported.

Fast, Reliable Head Tracking under Varying Illumination

An improved technique for 3D head tracking under varying illumination conditions is proposed. The head is modeled as a texture mapped cylinder. Tracking is formulated as an image registration problem in the cylinder's texture map image. To solve the registration problem in the presence of lighting variation and head motion, the residual error of registration is modeled as a linear combination of texture warping templates and orthogonal illumination templates. Fast and stable on-line tracking is then achieved via regularized, weighted least squares minimization of the registration error. The regularization term tends to limit potential ambiguities that arise in the warping and illumination templates. It enables stable tracking over extended sequences. Tracking does not require a precise initial fit of the model; the system is initialized automatically using a simple 2-D face detector. The only assumption is that the target is facing the camera in the first frame of the sequence. The warping templates are computed at the first frame of the sequence. Illumination templates are precomputed off-line over a training set of face images collected under varying lighting conditions. Experiments in tracking are reported.

Estimation and Prediction of Evolving Color Distributions for Skin Segmentation Under Varying Illumination

A novel approach for real-time skin segmentation in video sequences is described. The approach enables reliable skin segmentation despite wide variation in illumination during tracking. An explicit second order Markov model is used to predict evolution of the skin color (HSV) histogram over time. Histograms are dynamically updated based on feedback from the current segmentation and based on predictions of the Markov model. The evolution of the skin color distribution at each frame is parameterized by translation, scaling and rotation in color space. Consequent changes in geometric parameterization of the distribution are propagated by warping and re-sampling the histogram. The parameters of the discrete-time dynamic Markov model are estimated using Maximum Likelihood Estimation, and also evolve over time. Quantitative evaluation of the method was conducted on labeled ground-truth video sequences taken from popular movies.

Recursive Estimation of Motion and Planar Structure

A specialized formulation of Azarbayejani and Pentland's framework for recursive recovery of motion, structure and focal length from feature correspondences tracked through an image sequence is presented. The specialized formulation addresses the case where all tracked points lie on a plane. This planarity constraint reduces the dimension of the original state vector, and consequently the number of feature points needed to estimate the state. Experiments with synthetic data and real imagery illustrate the system performance. The experiments confirm that the specialized formulation provides improved accuracy, stability to observation noise, and rate of convergence in estimation for the case where the tracked points lie on a plane.

Inferring Body Pose without Tracking Body Parts

A novel approach for estimating articulated body posture and motion from monocular video sequences is proposed. Human pose is defined as the instantaneous two dimensional configuration (i.e., the projection onto the image plane) of a single articulated body in terms of the position of a predetermined set of joints. First, statistical segmentation of the human bodies from the background is performed and low-level visual features are found given the segmented body shape. The goal is to be able to map these, generally low level, visual features to body configurations. The system estimates different mappings, each one with a specific cluster in the visual feature space. Given a set of body motion sequences for training, unsupervised clustering is obtained via the Expectation Maximation algorithm. Then, for each of the clusters, a function is estimated to build the mapping between low-level features to 3D pose. Currently this mapping is modeled by a neural network. Given new visual features, a mapping from each cluster is performed to yield a set of possible poses. From this set, the system selects the most likely pose given the learned probability distribution and the visual feature similarity between hypothesis and input. Performance of the proposed approach is characterized using a new set of known body postures, showing promising results.

Index Trees for Efficient Deformable Shape-based Retrieval

An improved method for deformable shape-based image indexing and retrieval is described. A pre-computed index tree is used to improve the speed of our previously reported on-line model fitting method; simple shape features are used as keys in a pre-generated index tree of model instances. In addition, a coarse to fine indexing scheme is used at different levels of the tree to further improve speed while maintaining matching accuracy. Experimental results show that the speedup is significant, while accuracy of shape-based indexing is maintained. A method for shape population-based retrieval is also described. The method allows query formulation based on the population distributions of shapes in each image. Results of population-based image queries for a database of blood cell micrographs are shown.

3D Hand Pose Reconstruction Using Specialized Mappings

A system for recovering 3D hand pose from monocular color sequences is proposed. The system employs a non-linear supervised learning framework, the specialized mappings architecture (SMA), to map image features to likely 3D hand poses. The SMA's fundamental components are a set of specialized forward mapping functions, and a single feedback matching function. The forward functions are estimated directly from training data, which in our case are examples of hand joint configurations and their corresponding visual features. The joint angle data in the training set is obtained via a CyberGlove, a glove with 22 sensors that monitor the angular motions of the palm and fingers. In training, the visual features are generated using a computer graphics module that renders the hand from arbitrary viewpoints given the 22 joint angles. We test our system both on synthetic sequences and on sequences taken with a color camera. The system automatically detects and tracks both hands of the user, calculates the appropriate features, and estimates the 3D hand joint angles from those features. Results are encouraging given the complexity of the task.

Region Segmentation via Deformable Model-Guided Split and Merge

An improved method for deformable shape-based image segmentation is described. Image regions are merged together and/or split apart, based on their agreement with an a priori distribution on the global deformation parameters for a shape template. The quality of a candidate region merging is evaluated by a cost measure that includes: homogeneity of image properties within the combined region, degree of overlap with a deformed shape model, and a deformation likelihood term. Perceptually-motivated criteria are used to determine where/how to split regions, based on the local shape properties of the region group's bounding contour. A globally consistent interpretation is determined in part by the minimum description length principle. Experiments show that the model-based splitting strategy yields a significant improvement in segmention over a method that uses merging alone.

Foreground Object Segmentation from Binocular Stereo Video

Moving cameras are needed for a wide range of applications in robotics, vehicle systems, surveillance, etc. However, many foreground object segmentation methods reported in the literature are unsuitable for such settings; these methods assume that the camera is fixed and the background changes slowly, and are inadequate for segmenting objects in video if there is significant motion of the camera or background. To address this shortcoming, a new method for segmenting foreground objects is proposed that utilizes binocular video. The method is demonstrated in the application of tracking and segmenting people in video who are approximately facing the binocular camera rig. Given a stereo image pair, the system first tries to find faces. Starting at each face, the region containing the person is grown by merging regions from an over-segmented color image. The disparity map is used to guide this merging process. The system has been implemented on a consumer-grade PC, and tested on video sequences of people indoors obtained from a moving camera rig. As can be expected, the proposed method works well in situations where other foreground-background segmentation methods typically fail. We believe that this superior performance is partly due to the use of object detection to guide region merging in disparity/color foreground segmentation, and partly due to the use of disparity information available with a binocular rig, in contrast with most previous methods that assumed monocular sequences.

Some Considerations on a Calculus with Weak References

Weak references are references that do not prevent the object they point to from being garbage collected. Most realistic languages, including Java, SML/NJ, and OCaml to name a few, have some facility for programming with weak references. Weak references are used in implementing idioms like memoizing functions and hash-consing in order to avoid potential memory leaks. However, the semantics of weak references in many languages are not clearly specified. Without a formal semantics for weak references it becomes impossible to prove the correctness of implementations making use of this feature. Previous work by Hallett and Kfoury extends λgc, a language for modeling garbage collection, to λweak, a similar language with weak references. Using this previously formalized semantics for weak references, we consider two issues related to well-behavedness of programs. Firstly, we provide a new, simpler proof of the well-behavedness of the syntactically restricted fragment of λweak defined previously. Secondly, we give a natural semantic criterion for well-behavedness much broader than the syntactic restriction, which is useful as principle for programming with weak references. Furthermore we extend the result, proved in previously of λgc, which allows one to use type-inference to collect some reachable objects that are never used. We prove that this result holds of our language, and we extend this result to allow the collection of weakly-referenced reachable garbage without incurring the computational overhead sometimes associated with collecting weak bindings (e.g. the need to recompute a memoized function). Lastly we use extend the semantic framework to model the key/value weak references found in Haskell and we prove the Haskell is semantics equivalent to a simpler semantics due to the lack of side-effects in our language.