The recent years have witnessed advances in parallel algorithms for large-scale optimization problems. Notwithstanding the demonstrated success, existing algorithms that parallelize over features are usually limited by divergence issues under high parallelism or require data preprocessing to alleviate these problems. In this paper, we propose a Parallel Coordinate Descent algorithm using approximate Newton steps (PCDN) that is guaranteed to converge globally without data preprocessing. The key component of the PCDN algorithm is the high-dimensional line search, which guarantees the global convergence with high parallelism. The PCDN algorithm randomly partitions the feature set into b subsets/bundles of size P, and sequentially processes each bundle by first computing the descent directions for each feature in parallel and then conducting P-dimensional line search to compute the step size. We show that: 1) the PCDN algorithm is guaranteed to converge globally despite increasing parallelism and 2) the PCDN algorithm converges to the specified accuracy ϵ within the limited iteration number of Tϵ, and Tϵdecreases with increasing parallelism. In addition, the data transfer and synchronization cost of the P-dimensional line search can be minimized by maintaining intermediate quantities. For concreteness, the proposed PCDN algorithm is applied to L1-regularized logistic regression and L1-regularized L2-loss support vector machine problems. Experimental evaluations on seven benchmark data sets show that the PCDN algorithm exploits parallelism well and outperforms the state-of-the-art methods.

As a scene-independent descriptor of crowd motions, crowd collectiveness quantifies the degree of constituent individuals moving as a union in a crowd scene. An effective measurement on crowd collectiveness is of great importance for applications in surveillance of public safety, human dynamics, and other areas. To this end, we propose a novel framework to measure crowd collectiveness by combining macroscopic and microscopic motion consistencies and define quantitatively the global and local consistency of crowd motions. The defined global consistency represents the likelihood of pairwise individuals belonging to the same collective group, whereas the local consistency reflects the degree of conformity in a local region. Based on the proposed collectiveness measure, a new algorithm, named group mining, is proposed to detect collective groups from a crowd. We validate the effectiveness of the proposed method on several synthetic particle systems and a real-world crowd database with human-labeled collectiveness. Experimental results show that, compared with the previous approaches, our collectiveness measure is more consistent with human perception, and the collective groups detected by our group mining algorithm are more accurate and robust.

Human activity recognition is a challenging high-level vision task, for which multiple factors, such as subject, object, and their diverse interactions, have to be considered and modeled. Current learning-based methods are limited in the capability to integrate human-level concepts into an easily extensible computational framework. Inspired by the existing human memory model, we present a context-associative approach to recognize activity with human-object interaction. The proposed system can recognize incoming visual content based on the previous experienced activities. The high-level activity is parsed into consecutive subactivities, and we build a context cluster to model the temporal relations. The semantic attributes of the subactivity are organized by a concept hierarchy. Based on the hierarchy, a series of similarity functions are defined to turn the recognition computing into retrievals over the contextual memory, similar to the auto-associative characteristics of human memory. Partially matching in retrieval and stored memory make the activity prediction possible. The dynamical evolution of the brain memory is mimicked to allow decay and reinforcement of the input information, providing a natural way to maintain data and save computational time. We evaluate our approach on three data sets: CAD-120, MHOI, and OPPORTUNITY. The proposed method demonstrates promising results compared with other state-of-the-art techniques.

We address the problem of pose estimation in videos. The part detectors play important roles, but traditional template-based detectors (e.g. Histogram of Gradient, HoG) fail at pose estimation due to the high variability in appearance. We present an adaptive representation of appearance and shape for articulated human body. The full representation of human body is based on the flexible mixture-of-parts model. We train a Naive Bayes classifier to obtain a confidence score of estimated pose by the basic mixture model, and based on the confidence we learn an instance-specific appearance model. For between-frame consistency, we design a time-efficient energy function for motion cues instead of complex motion models. We incorporate these models into a framework that allows for efficient inference. Quantitative evaluation of pose estimation conducted on two video datasets demonstrates the effectiveness of the proposed method.

In this work, we propose a discriminative tracklets representation for motion pattern extraction from crowded scene. The representation incorporates relative position, velocity, and direction information of tracklet into one compact form by shaping it within a rectangle. We adopt deep belief networks to extract low-dimensional features from this representation. It not only reduces the computational complexity for the following clustering, but also achieves more discriminative tracklets representation which is invariant to noises brought by tracking failures. To determine the spatio-temporal distribution of each motion pattern, a robust clustering scheme composed of three clustering procedures is proposed. Comprehensive experiments in multiple datasets validate the effectiveness of our approach.

Coherent motion is a very common motion pattern in crowded scenes. Coherent Filter is a very effective and robust tool to detect coherent motions based on point trajectories, the performance of coherent filter depends on point trajectories' property. In this work, we present a two-stage strategy to extract dense, accurate and long-term point trajectories from crowded scenes. The method includes a tracklets acquisition procedure and a tracklets association procedure. We use LDOF tracker to acquire dense tracklets, and then formulate tracklets association as a linear assignment problem (LAP). Experiments conducted on challenging crowd datasets show that our trajectories are very suitable for detecting coherent motions in crowded scenes.

Classical methods recognise vehicle logos with image feature matching approaches. Different from these methods, this study proposes a novel algorithm to recognise the vehicle logos in real time by constructing the weighted multi‐class support vector machine (SVM) ensemble model to classify the vehicle logos based on sharpness histogram features. To evaluate the performance of the proposed algorithm, extensive experiments have been performed. Experimental results indicate that the sharpness histogram features proposed by the authors has better distinguishability than colour histogram features. Moreover, they show that the proposed algorithm has the best average recognition performance, and its performance is the most robust. Conveniently, the proposed algorithm can avoid the burden of choosing the appropriate kernel function and parameters comparing with multi‐class SVM model.

Constrained spectral clustering (CSC) algorithms have shown great promise in significantly improving clustering accuracy by encoding side information into spectral clustering algorithms. However, existing CSC algorithms are inefficient in handling moderate and large datasets. In this paper, we aim to develop a scalable and efficient CSC algorithm by integrating sparse coding based graph construction into a framework called constrained normalized cuts. To this end, we formulate a scalable constrained normalized-cuts problem and solve it based on a closed-form mathematical analysis. We demonstrate that this problem can be reduced to a generalized eigenvalue problem that can be solved very efficiently. We also describe a principled k-way CSC algorithm for handling moderate and large datasets. Experimental results over benchmark datasets demonstrate that the proposed algorithm is greatly cost-effective, in the sense that (1) with less side information, it can obtain significant improvements in accuracy compared to the unsupervised baseline; (2) with less computational time, it can achieve high clustering accuracies close to those of the state-of-the-art.

Person re-identification is an important and challenging computer vision problem. Recent progress in this area is due to new visual features and models that deals with cross-view variations. Instead of working towards more complex models, we focus on low level features and their encoding. Low level features capturing the color and structural information are first extracted from human images. Gaussian Mixture Model (GMM) is then employed to approximate the distribution of the features, providing a relatively comprehensive statistical representation. Finally, low level features are mapped to a space by computing free energy score of the GMM. The mapped features are encoded into a fixed-length feature vector for person re-identification. Extensive experiments are conducted on several public datasets. Comparisons with benchmark person re-identification methods show the promising performance of our approach.

This paper presents a study of an important imaging system, called the plate refractive imaging system. The system comprises of a perspective camera looking through one or more planar parallel plates to the target imaging environment. Such as the perspective camera captures the object through an infrared filter which is used in the infrared navigator for surgeries. Sometimes, we have to calibrate the camera without removing the plate. Up to now, it does not have a valid algorithm to do so. In this paper, a plate refractive camera calibration model is proposed for the imaging systems that incorporate a refractive plate at the front end of the camera. The proposed model is based on the refractive effect of variations in the viewpoint of such systems. Both synthetic and real data show that our proposed camera model is insensitive to noise and can achieve high accuracy.

Current research on traffic flow prediction mainly concentrates on generating accurate prediction results based on intelligent or combined algorithms but ignores the interpretability of the prediction model. In practice, however, the interpretability of the model is equally important for traffic managers to realize which road segment in the road network will affect the future traffic state of the target segment in a specific time interval and when such an influence is expected to happen. In this paper, an interpretable and adaptable spatiotemporal Bayesian multivariate adaptive-regression splines (ST-BMARS) model is developed to predict short-term freeway traffic flow accurately. The parameters in the model are estimated in the way of Bayesian inference, and the optimal models are obtained using a Markov chain Monte Carlo (MCMC) simulation. In order to investigate the spatial relationship of the freeway traffic flow, all of the road segments on the freeway are taken into account for the traffic prediction of the target road segment. In our experiments, actual traffic data collected from a series of observation stations along freeway Interstate 205 in Portland, OR, USA, are used to evaluate the performance of the model. Experimental results indicate that the proposed interpretable ST-BMARS model is robust and can generate superior prediction accuracy in contrast with the temporal MARS model, the parametric model autoregressive integrated moving averaging (ARIMA), the state-of-the-art seasonal ARIMA model, and the kernel method support vector regression.

Learning the content-level similarity over images is a fundamental problem in content-based image retrieval, and is highly challenging due to the variance within images. Such variance requires the similarity measure to be adaptive enough. In this paper, we propose a similarity learning approach based on the probabilistic modeling of images. First, we derive the similarity measure, free energy score space kernel (FESS kernel), from the probabilistic models. FESS kernel is essentially the function of the observed data, hidden variable and model parameters, where the hidden variables are very informative and are absent in previous methods. Then, we propose a discriminative learning approach for FESS kernel, encouraging the similarity to take a large value for the image pair with the same label and to take a small value for the image pair with distinct labels. The learned similarity fully exploits the data distribution and class label, and inherits the adaptability. We evaluate the proposed method on three databases. The results validate its competitive performance.

Crowded scene analysis is becoming increasingly popular in computer vision field. In this paper, we propose a novel approach to analyze motion patterns by clustering the hybrid generative-discriminative feature maps using unsupervised hierarchical clustering algorithm. The hybrid generative-discriminative feature maps are derived by posterior divergence based on the tracklets which are captured by tracking dense points with three effective rules. The feature maps effectively associate low-level features with the semantical motion patterns by exploiting the hidden information in crowded scenes. Motion pattern analyzing is implemented in a completely unsupervised way and the feature maps are clustered automatically through hierarchical clustering algorithm building on the basis of graphic model. The experiment results precisely reveal the distributions of motion patterns in current crowded videos and demonstrate the effectiveness of our approach.

Accurate short-term traffic flow prediction plays a fundamental role in intelligent transportation systems (ITS), e.g. advanced traffic management systems (ATMS). To generate accurate short-term traffic volume, nonparametric models have gained credit from quantities of researchers. On the basis of the common thought that future traffic states can be predicted according to the similar states in the historical traffic data, this paper presents a novel nonparametric-model-based method to predict the short-term traffic volume. The applied nonparametric model is the classification and regression trees (CART) model. In the application, the CART model first classifies the historical traffic states into plentiful categories. Afterwards, the linear regression model is built corresponding to each traffic state pattern. Finally, the model predicts the short-term traffic state through clustering the current state vector into the most congenial historical pattern and regression model. In the experiments, the proposed method is tested by using the 15 minutes average traffic volumes on freeways and is compared with the classic nonparametric methods k-nearest neighbors (k-NN) model, and the parametric method Kalman filter model. The results indicate that the CART-based prediction method outperforms the k-NN and Kalman filter methods in both the mean absolute percentage error and the mean absolute scaled error.

Articulated configuration of human body parts is an essential representation of human motion, therefore is well suited for classifying human actions. In this work, we propose a novel approach to exploring the discriminative pose sub-patterns for effective action classification. These pose sub-patterns are extracted from a predefined set of 3D poses represented by hierarchical motion angles. The basic idea is motivated by the two observations: (1) There exist representative sub-patterns in each action class, from which the action class can be easily differentiated. (2) These sub-patterns frequently appear in the action class. By constructing a connection between frequent sub-patterns and the discriminative measure, we develop the SSPI, namely, the Support Sub-Pattern Induced learning algorithm for simultaneous feature selection and feature learning. Based on the algorithm, discriminative pose sub-patterns can be identified and used as a series of "magnetic centers" on the surface of normalized super-sphere for feature transform. The "attractive forces" from the sub-patterns determine the direction and step-length of the transform. This transformation makes a feature more discriminative while maintaining dimensionality invariance. Comprehensive experimental studies conducted on a large scale motion capture dataset demonstrate the effectiveness of the proposed approach for action classification and the superior performance over the state-of-the-art techniques.

Small sample set, occlusion, and illumination variations are the critical obstacles for a face identification system towards practical application. In this paper, we propose a probabilistic generative model for parts-based data representation to address these difficulties. In our approach, multiple subcategories corresponding to the individual face parts, such as nose, mouth, eye, and so forth, are modeled within a probabilistic graphical model framework to mimic the process of generating a face image. The induced face representation, therefore, encodes rich discriminative information. Model training is totally unsupervised. Once the training is completed, a test sample from the face class can be recognized as a novel combination of learned parts. In summary, the main contributions of this work are threefold: 1) A novel hierarchical probabilistic generative model is proposed, which is capable of achieving an efficient parts-based representation for robust face identification. 2) A constrained variational EM algorithm is developed to learn the model parameters and infer the variables. 3) Two similarity metrics are specially designed for the novel parts-based feature representation, which are effective for matching score guided one sample face identification. The models and similarity metrics are validated on three face databases. Experimental results demonstrate the capabilities of the model to deal with small sample set, occlusions, and illumination variances.

Aiming to comply with the requirement of parallel-transportation management systems (PtMS), this paper presents a short-term traffic flow prediction method for signal-controlled urban traffic networks (UTNs) based on the macroscopic UTN model. In contrast with other time-series-based or spatio-temporal correlation methods, the proposed method focuses more on using the substantial mechanism of traffic transmission in road networks and the topology model of the entire UTN. Furthermore, this approach employs a speed-density model based on the fundamental diagram (FD) to obtain more accurate travel times in links. In the comparison experiment, the microscopic traffic simulation software CORSIM is adopted to simulate the real urban traffic. The experiment results fully verify the outstanding performances of the proposed prediction method.

This paper presents a systematic solution to efficiently estimate the traffic state of large-scale urban road networks. We first propose the new approach to construct the exact GIS-T digital map. The exact digital map can lay the solid foundation for the traffic state estimation with the data from Global Positioning System (GPS) probe vehicles. Then, we present the following two effective methods based on GPS probe vehicles for the traffic state estimation: (1) the curve-fitting-based method and (2) the vehicle-tracking-based method. Finally, we test the proposed solution with a large number of real data from GPS probe vehicles and the standard digital map of Shanghai, China. In the experiments, data from thousands of GPS-equipped taxies were taken as the probe vehicles. The estimation accuracy and operation speed of the two different methods were systematically measured and compared. In addition, the coverages of the GPS sampling points were also investigated for the large-scale urban road network in the spatial and temporal domains. For the accuracy experiment, the ground truth was obtained by repeating the videos that were recorded on 24 road sections in downtown Shanghai. The experimental results illustrate that the proposed methods are effective and efficient in monitoring the traffic state of large-scale urban road networks.

The overwhelming amounts of digital images on the Web and personal computers have triggered the requirement of an effective tool to retrieve images of interest using semantic concepts. Due to the semantic gap between low level content features and its high level semantic features of an image, however, the performances of many existing automatic image annotation algorithms are not so satisfactory. In this paper, a novel approach based on the cognitive science is proposed to improve the quality of annotations. The main idea is that the tags of an image are considered as nodes in a semantic network, and the relevance between the tags and image contents is regulated using the spreading activation theory. After the spreading activation process finishes, each tag will be assigned an appropriate value with respect to its relation to other tags. Experimental results conducted on the 50,000 Flickr images demonstrate that the proposed scheme can effectively improve the performance in automatic image annotation.

Recently learning to rank has become one of the popular means to create a ranking model for social image search. However, the results of existing approaches are not as satisfactory for the large gap between low-level visual features and high-level semantic concepts, and these sophisticated approaches require a significant amount of parameters tuning to be effective and efficient. In this paper, we propose a novel framework for social image re-ranking based on a non-parametric quantum technique, which reranks top retrieved images by considering the interrelationship between images through the quantum estimation and requires no explicit parameter tuning. The basic idea of the proposed framework is inspired by the photon polarization experiment supporting the theory of quantum measurement. Experimental results conducted on the Flickr dataset demonstrate the effectiveness and efficiency of the proposed framework.

We propose a novel human action recognition method based on the generative feature mapping over 3D human body joint sequences. The proposed method relies on Hidden Markov Model (HMM), but differs from the previous methods in the way of incorporating HMM and discriminative classifier, aiming to capture more discriminative information. Firstly, we use HMMs to model the joint sequences of human body. Then the Posterior Divergence is used to build feature mappings from the trained HMMs. The derived feature mappings map a variable-length joint sequence to a fixed-dimension feature vector which will be delivered to SVM for classification. We evaluate the proposed method and related methods on a large number of 3D joint sequences. The experimental results show its competitive performance, in comparison with other state-of-the-art methods.

Integrating generative models and discriminative models in a hybrid scheme has shown some success in recognition tasks. In such scheme, generative models are used to derive feature maps for outputting a set of fixed length features that are used by discriminative models to perform classification. In this paper, we present a method, called posterior divergence, to derive feature maps from the log likelihood function implied in the incremental expectation-maximization algorithm. These feature maps evaluate a sample in three complementary measures: (1) how much the sample affects the model; (2) how well the sample fits the model; (3) how uncertain the fit is. We prove that the linear classification error rate using the outputs of the derived feature maps is at least as low as that of plug-in estimation. We present efficient algorithms for computing these feature maps for semi-supervised learning and supervised learning. We evaluate the proposed method on three typical applications, i.e. scene recognition, face and non-face classification and protein sequence analysis, and demonstrate improvements over related methods.

Background reconstruction plays an important role in many applications like video surveillance, motion analysis. Traditional Adaptive Gaussian Mixture Model will lose target when deal with arbitrary-long stationary object. In this paper, a novel method for detecting this kind of object is proposed to improve the performance of Adaptive Gaussian Mixture Model. Parameter restoration is designed to deal with arbitrary-long stationary target and solve the short-comings of the latest algorithm. The parameters among the K distribution of each pixel covered by the object will be restored when the object stayed for over threshold frames. Then the target will not be updated as a part of background model. Experimental results show that the proposed algorithm proves to be a more robust method by detecting the stationary target in an arbitrary-long time.

This paper proposes a method for background modeling and foreground detection in video. This method divides the background into two layers, the dynamic layer and the static layer. An energy descriptor is proposed to analysis the motion state in dynamic layer while a grid filter is proposed to reduce the negative impact of sudden illumination change such as light switching off. Experiment results compared with four typical algorithms show that this method outperforms others in most challenging videos including sudden illumination change and some complex backgrounds.

Detecting motion pattern in dynamic crowd scenes is a challenging problem in computer vision field. In this paper, we propose a novel approach to detect the motion patterns from global perspective. To extract the discriminative spatial-temporal features, we introduce the Motion History Image (MHI) into the optical flow algorithm. Motion patterns are then detected by automatic clustering of optical flow vectors through hierarchical clustering. Experiment evaluation on some challenging videos shows reliable detection results and demonstrates the effectiveness of our proposed approach.

The problem of real-time rectangle detection on high-resolution image arises in actual panel production. This paper proposes a robust real-time method for panel rectangle detection. In line extraction part, a new technique called "Crossing Dilation" is proposed to improve the Progressive Probabilistic Hough Transform performance, to extract long line segments that are not perfectly straight in large sparse binary image. The rectangle is detected using the proposed rectangle fitting method. Experiments on a variety of real cases of large panels are executed, some of which are badly disturbed by the fake edges caused by the dust in the background region. The results show great real-time performance on all the cases, and the proposed method is robust to the disturbances and different panel sizes.

This paper presents a novel method for detection and recognition of glass defects in low resolution images. First, the defect region is located by the method of Canny edge detection, and thus the smallest connected region (rectangle) can be found. Then, the binary information of the core region can be obtained based on a specific filter. After noises are removed, a novel Binary Feature Histogram (BFH) is proposed to describe the characteristic of the glass defect. Finally, the AdaBoost method is adopted for classification. The classifiers are designed based on BFH. Experiments with 800 bubble images and 240 non-bubble images prove that the proposed method is effective and efficient for recognition of glass defects, such as bubbles and inclusions.

Existing approaches for automatic image annotation usually suffer from two issues: (1) lacking a good quality distance metric for image semantic similarity measure; (2) rarely considering the correlation between labels assigned to each image. In this paper, we aim to resolve both of the problems simultaneously in a novel unified framework. Specifically, a proper distance metric is learned based on the structural SVM in a discriminative manner, which can optimize the ranking of the images induced by distances from a test image. Subsequently, a collaborative label propagation algorithm is leveraged to model the correlation between class labels in an explicit manner. Also, the learned metric is embedded in the propagation model. The integration of the two components leads to more accurate annotation results. The experiments conducted on the Corel dataset demonstrate the effectiveness of the proposed unified framework.

Human pose estimation via motion tracking systems can be considered as a regression problem within a discriminative framework. It is always a challenging task to model the mapping from observation space to state space because of the high-dimensional characteristic in the multimodal conditional distribution. In order to build the mapping, existing techniques usually involve a large set of training samples in the learning process which are limited in their capability to deal with multimodality. We propose, in this work, a novel online sparse Gaussian Process (GP) regression model to recover 3-D human motion in monocular videos. Particularly, we investigate the fact that for a given test input, its output is mainly determined by the training samples potentially residing in its local neighborhood and defined in the unified input-output space. This leads to a local mixture GP experts system composed of different local GP experts, each of which dominates a mapping behavior with the specific covariance function adapting to a local region. To handle the multimodality, we combine both temporal and spatial information therefore to obtain two categories of local experts. The temporal and spatial experts are integrated into a seamless hybrid system, which is automatically self-initialized and robust for visual tracking of nonlinear human motion. Learning and inference are extremely efficient as all the local experts are defined online within very small neighborhoods. Extensive experiments on two real-world databases, HumanEva and PEAR, demonstrate the effectiveness of our proposed model, which significantly improve the performance of existing models.

Detecting text and caption from videos is important and in great demand for video retrieval, annotation, indexing, and content analysis. In this paper, we present a corner based approach to detect text and caption from videos. This approach is inspired by the observation that there exist dense and orderly presences of corner points in characters, especially in text and caption. We use several discriminative features to describe the text regions formed by the corner points. The usage of these features is in a flexible manner, thus, can be adapted to different applications. Language independence is an important advantage of the proposed method. Moreover, based upon the text features, we further develop a novel algorithm to detect moving captions in videos. In the algorithm, the motion features, extracted by optical flow, are combined with text features to detect the moving caption patterns. The decision tree is adopted to learn the classification criteria. Experiments conducted on a large volume of real video shots demonstrate the efficiency and robustness of our proposed approaches and the real-world system. Our text and caption detection system was recently highlighted in a worldwide multimedia retrieval competition, Star Challenge, by achieving the superior performance with the top ranking.

This paper focuses on multimodal gender recognition. To achieve a robust and discriminative performance for gender recognition, visual observations from both face and corresponding fingerprints are fused to serve for the task. The bag-of-words model is employed to structure the image representation. We propose a novel supervised method to construct the visual words, by which the redundant feature dimensions are discarded and the important dimensions for gender classification are highlighted. The dimension rearrangement is achieved by aligning the feature dimensions to a common normal vector of the hyperplane between categories. The Latent Dirichlet Allocation (LDA) model is extended to incorporate discriminative clues for supervised classification. We build the novel Discriminative LDA (D-LDA) model by maximizing the inter-class margins, which can significantly enhance the discriminative power of the whole model. Experiments on a large face and fingerprint database demonstrate the effectiveness of the proposed new feature and model. Complementary advantages benefited from face-fingerprint fusion to a robust gender recognition framework also get validated.

We propose to estimate human gender from corresponding fingerprint and face information with the Bayesian hierarchical model. Different from previous works on fingerprint based gender estimation with specially designed features, our method extends to use general local image features. Furthermore, a novel word representation called latent word is designed to work with the Bayesian hierarchical model. The feature representation is embedded to our multimodality model, within which the information from fingerprint and face is fused at the decision level for gender estimation. Experiments on our internal database show the promising performance.

This paper addresses a strategy for 3D human motion recovery from monocular image. We advocate the use of Gaussian Process Dynamical Model (GPDM) for learning human pose and motion priors for 3D people tracking. With the prior learned from GPDM, we integrate our approach into a Bayesian tracking framework of condensation. During the off-line training step, a GPDM provides the reversible mappings between low-dimensional latent space and high-dimensional pose space, and then in the online tracking process, the latent variables are estimated via the particle filtering, and the observation is designed as a energy function based on a Markov Random Field (MRF) theory. The proposed approach is demonstrated on our database, and the experimental results show that our method performs promisingly.

Saliency mechanism has been considered crucial in the human visual system and helpful to object detection and recognition. This paper addresses a novel feature-based model for visual saliency detection. It consists of two steps: first, using the learned overcomplete sparse bases to represent image patches; and then, estimating saliency information via direct low-rank and sparsity matrix decomposition. We compare our model with the previous methods on natural images. Experimental results show that our model performs competitively for visual saliency detection task, and suggest the potential application of matrix decomposition and convex optimization for image analysis.

This paper addresses a method for 3D human motion tracking and voxel-based reconstruction from sparse views. We adopt the annealed Gaussian based particle swarm optimization (AGPSO) for 3D human motion tracking. The AGPSO algorithm incorporates the temporal continuity information into the traditional particle swarm optimization (PSO) algorithm under a Bayesian framework. In the online tracking process, the state variables are estimated via the particle filtering, where the observation is designed as a minimized Markov Random Field (MRF) energy. Finally, voxel reconstruction is conducted using the skeleton shape prior via dynamic graph cut. The experimental results show that our method performs promisingly against the cluttered background and generates plausible voxel reconstructions from sparse views.

We present a method for 3D shape reconstruction of inextensible deformable surfaces from monocular image sequences. The key of our approach is to represent the surface as 3D triangulated mesh and formulate the reconstruction problem as a sequence of Linear Programming (LP) problems which can be effectively solved. The LP problem consists of data constraints which are 3D-to-2D keypoint correspondences and shape constraints which prevent large changes of the edge orientation between consecutive frames. Furthermore, we use a refined bisection algorithm to accelerate the computing speed. The robustness and efficiency of our approach are validated on both synthetic and real data.

This paper proposes to learn parts-based face representation from real face samples and then applies it to face transition. It differs from previous works in two aspects. First, we learn flexible face decomposition from real faces unsupervisedly instead of designing face template manually, for which two simple priors are embedded into learning procedure through constrained EM formulation. Second, both face representation and transition are derived from an unified probabilistic framework. Based on the learned face representation, the face distance measurement could be defined, which enables us to synthesize face via specifying distance with respect to reference faces and depict the full transition trace of two or more given faces with distinct age, gender and race.

A new method of image based flow analysis and synthesis from a still image is presented. We analyze the flow parts from a still image in order to get continuous video by image matting, image inpainting, projecting flow field onto the image and modulation. We construct and project 3D flow models onto still images and propose 2D modulation methods that are more suitable and practical to our synthesis. Our technique can edit still images that have flow parts. The experiments show the method is effective.

In computer vision community, human pose estimation and nonrigid shape recovery have evolved into different subfields. The state-of-the-art optimization techniques have been applied to the problem of deformable surface reconstruction successfully and recent methods in this area have focused on designing formulations that are easier to solve. In general, these techniques lay their success on the assumption that sufficient 2-D-3-D correspondences can be detected. By contrast, confronted with the similar ambiguity problem, many techniques for human pose estimation adopt stochastic searching or discriminative predictions, which allow for more generative image cues. However, the global optimization cannot be guaranteed via the stochastic methods; and discriminative techniques usually suffer from inaccuracy. In this letter, we absorb ideas from both domains and propose a unified approach for articulated human pose estimation. Specifically, we optimize the human pose to account for the discriminative pose prediction, bone length preservation in parallel with the point-topoint image observation. Moreover, the L2norm minimization is solved iteratively as a linear system with high computational efficiency.

Saliency mechanism has been considered crucial in the human visual system and helpful to object detection and recognition. This paper addresses a novel feature-based model for visual saliency detection. It consists of two steps: first, using the learned overcomplete sparse bases to represent image patches; and then, estimating saliency information via low-rank and sparsity matrix decomposition. We compare our model with the previous methods on natural images. Experimental results on both natural images and psychological patterns show that our model performs competitively for visual saliency detection task, and suggest the potential application of matrix decomposition and convex optimization for image analysis.

We consider the problem of estimating 3-D human body pose from visual signals within a discriminative framework. It is challenging because there is a wide gap between complex 3-D human motion and planar visual observation, which makes this a severely ill-conditioned problem. In this paper, we focus on three critical factors to tackle human body pose estimation, namely, feature extraction, learning algorithm, and camera utilization. On the feature level, we describe images using the salient interest points represented by scale-invariant feature transform (SIFT)-like descriptors, in which the position, appearance, and local structural information are encoded simultaneously. On the learning algorithm level, we propose to use Gaussian processes and multiple linear (ML) regression to model the mapping between poses and features. Fusing image information from multiple cameras in different views is of great interest to us on the camera level. We make a comprehensive evaluation on the HumanEva database and get two meaningful insights into the three crucial aspects for human pose estimation: 1) although the choice of feature is very important to the problem, once the learning algorithm becomes efficient, the choice of feature is no longer critical, and 2) the impact of information combination from multiple cameras on pose estimation is closely related to not only the quantity of image information, but also its quality. In most cases, it is true that the more information is involved, the better results can be achieved. But when the information quantity is the same, the differences in quality will lead to totally different performance. Furthermore, dense evaluations demonstrate that our approach is an accurate and robust solution to the human body pose estimation problem.

We present a method for recovering 3-D nonrigid structure from an image pair taken with a stereo rig. More specifically, we dedicate to recover shapes of nearly inextensible deformable surfaces. In our approach, we represent the surface as a 3-D triangulated mesh and formulate the reconstruction problem as an optimization problem consisting of data terms and shape terms. The data terms are model to image keypoint correspondences which can be formulated as second-order cone programming (SOCP) constraints using L(infinity) norm. The shape terms are designed to retaining original lengths of mesh edges which are typically nonconvex constraints. We will show that this optimization problem can be turned into a sequence of SOCP feasibility problems in which the nonconvex constraints are approximated as a set of convex constraints. Thanks to the efficient SOCP solver, the reconstruction problem can then be solved reliably and efficiently. As opposed to previous methods, ours neither involves smoothness constraints nor need an initial estimation, which enables us to recover shapes of surfaces with smooth, sharp and other complex deformations from a single image pair. The robustness and accuracy of our approach are evaluated quantitatively on synthetic data and qualitatively on real data.

We present a method for recovering the 3-D shape of an inextensible deformable surface from a monocular image sequence. State-of-the-art methods on this problem , utilize L(infinity)-norm of reprojection residual vectors and formulate the tracking problem as a Second-Order Cone Programming (SOCP) problem. Instead of using L(infinity) which is sensitive to outliers, we use L(2)-norm of reprojection errors. Generally, using L(2) leads a nonconvex optimization problem which is difficult to minimize. Instead of solving the nonconvex problem directly, we design an iterative L(2)-norm approximation process to approximate the nonconvex objective function, in which only a linear system needs to be solved at each iteration. Furthermore, we introduce a shape regularization term into this iterative process in order to keep the inextensibility of the recovered mesh. Compared with previous methods, ours performs more robust to image noises, outliers and large interframe motions with high computational efficiency. The robustness and accuracy of our approach are evaluated quantitatively on synthetic data and qualitatively on real data.

Accurate traffic parameters such as traffic flow, travel speeds and occupancies, are crucial to effective management of intelligent transportation systems (ITS). Some traffic data from loop detectors settled in arterial streets are incomplete, and the importance of effectively imputing the missing values emerges. In this paper, a technique called least squares support vector machines (LS-SVMs) is introduced to predict missing traffic flow based on spatio-temporal analysis in urban arterial streets. To the best of our knowledge, it is the first time to apply the rising computational intelligence (CI) technique incorporating with state space approach in missing traffic data imputation. Having good generalization ability and guaranteeing global minima ensure its well performance in the area. A baseline imputation technique, expectation maximization/data augmentation (EM/DA), is selected for comparison because of its proved effectiveness in missing data recovery. Through persuasive comparisons of the techniques, the proposed model is proved to be more applicable and performs better in stability and adaptability, which reveals that it is a promising approach in missing data prediction.

We consider the problem of 3D modeling under the environments where colors of the foreground objects are similar to the background, which poses a difficult problem of foreground and background classification. A purely image-based algorithm is adopted in this paper, with no prior information about the foreground objects. We classify foreground and background by fusing the information at the pixel and region levels to obtain the similarity probability map, followed by a Bayesian sensor fusion framework to infer the space occupancy grid. The estimation of the occupancy allows incremental updating once a new observation is available, and the contribution of each observation can be adjusted according to its reliability. Finally, three parameters in the algorithm are analyzed in detail and experiments show the effectiveness of this method.

We address the issue of markless human motion capture by voxel labeling. We explore the problem of pose estimation from voxel cloud based on a predefined human model. First, voxels are labeled into different individual parts of the body. Then, the joints are extracted from labeled voxels. Finally, the joint angles are estimated from the joints. Tested on the voxel data in our experiments, our algorithm is insensitive to noise and achieve fine accuracy in pose estimation.

We address the issue of categorizing scenes from feature films into semantic classifications based on the audio-visual cues. Specifically, we first exploit the grammar of film production to specify the semantic content of scenes. Then, each scene is classified into one of the following categories: conversation, action and suspense. Finally, to achieve more specific scene and consist with human perception, conversation scene is further categorizes into emotional conversation and common one, and action scene is further categorizes into gunfight, beating and chasing scene. This work is a step toward browsing and retrieval content of feature films in limited bandwidth, video repository, and rating of feature films of interest effectively and efficiently.

A novel approach of pose estimation is proposed for the object with surface of revolution(SOR). The silhouette of the object is the only information necessary for this method and no cross section circle (latitude circle) is needed. In this article, we explain the property of tangent circle and use it to establish constraint between two images of object with different poses. Such constraint can help to solve the pose of object in both images. We test our method with a simulation experiment and use it to estimate the pose for both rigid body and articulated object.

Graph-based semi-supervised learning approaches have been proven effective and efficient in solving the problem of the inefficiency of training samples in many real-world application areas, such as video annotation. As a significant factor of these algorithms, however, pairwise similarity metric has not been fully investigated. On the one hand, for general video annotation methods, the estimation of pairwise similarity between two samples relies on the spatial property of video data. On the other hand, temporal property, which is an essential character of video data, is not embedded into the pairwise similarity metric. Therefore, in this paper, a novel method, called Joint Spatio-Temporal Correlation Learning (JSTCL), is proposed to improve the accuracy of video annotation. This method is characterized by simultaneously taking into account both the spatial and temporal property of video data to well represent the pairwise similarity. Experiments conducted on the TRECVID demonstrate the efficiency of the proposed method.

Human matching between different fields of view is a difficult problem in intelligent video surveillance; whereas fusing multiple features has become a strong tool to solve it. In order to guide the fusion scheme, it is necessary to evaluate the matching performance of these features. In this paper, four typical features are chosen for the evaluation. They are the color histogram, UV chromaticity, major color spectrum histogram, and scale-invariant features (SIFT). Quantities of video data are collected to test their general accuracy, robustness, and real-time applicability. The robustness is measured under the conditions of illumination changes, Gaussian and salt noises,foreground errors, resolution changes, and camera angle differences. The experimental results show that the four features bear distinctive performances under the different conditions, which will provide important references for the feature fusion methods.

Natural phenomena such as smoke, waterfall, and stream show both texture and motion changes over time. The underlying physical properties bring complicated 3D variations which have attracted our interest. Velocity is one of important physical properties, which affects fluidity in small or large scale. Thus, this paper presents a 3D photodynamic tool (PDT) using a single fluid-like 2D image with 3D velocity primitives and physical models. Our tool offers only a few manipulations to generate a long term animation from a single image in a minute. Experimental results show that our 3D PDT can animate a realistic texture change from a single 2D photo even with occlusion in a scene.

This paper is about a vision-based system that automatically monitors intermodal freight trains for the quality of how the loads (containers) are placed along the train. An accurate and robust algorithm to segment the foreground of containers in videos of the moving train is indispensable for this purpose. Given a video of a moving train consisting of containers of different types, this paper presents a method exploiting the information in both frequency and spatial domains to segment these containers. This method can accurately segment all types of containers under a variety of background conditions, e.g illumination variations and moving clouds, in the train videos shot by a fixed camera. The accuracy and robustness of the proposed method are substantiated through a large number of experiments on real data of train videos.

Some traffic data from loop detectors settled in urban arterial streets are incomplete. The importance of effectively imputing the missing values emerges. The letter introduces least squares support vector machines (LS-SVMs) to missing traffic flow prediction based on spatio-temporal analysis. It is the first time to apply the technique to missing data imputation. A baseline imputation technique, expectation maximization/data augmentation (EM/DA), is selected for comparison because of its proved effectiveness. Experimental results demonstrate that our method is more applicable and performs better at relatively high missing data rates. This reveals that it is a promising approach in the field.

Automatic image annotation is a promising way to achieve more effective image management and retrieval. However, system performances of the existing state-of-the-art keyword annotation schemes are often not so satisfactory. Therefore, image annotation refinement is crucial to improve the imprecise annotation results. In this paper, a novel approach is developed to automatically annotate image content by a semi-supervised learning model. With perceptual visual characteristics, the candidate annotations of unlabelled images are first obtained based on a progressive model. Then, a transducitive model, random walk with restart algorithm is used to refine these candidate annotations and the top ones are reserved as the final annotations. Experiments conducted on the typical Corel dataset show the effectiveness of the proposed approach.

This paper presents an information-fusion-based approach to the estimation of urban traffic states. The approach can fuse online data from underground loop detectors and global positioning system (GPS)-equipped probe vehicles to more accurately and completely obtain traffic state estimation than using either of them alone. In this approach, three parts of the algorithms are developed for fusion computing and the data processing of loop detectors and GPS probe vehicles. First, a fusion algorithm, which integrates the federated Kalman filter and evidence theory (ET), is proposed to prepare a robust, credible, and extensible fusion platform for the fusion of multisensor data. After that, a novel algorithm based on the traffic wave theory is employed to estimate the link mean speed using single-loop detectors buried at the end of links. With the GPS data, a series of technologies are combined with the geographic information systems for transportation (GIS-T) map to compute another link mean speed. These two speeds are taken as the inputs of the proposed fusion platform. Finally, tests on the accuracy, conflict resistance, robustness, and operation speed by real-world traffic data illustrate that the proposed approach can well be used in urban traffic applications on a large scale.

A novel evolutionary algorithm called Probability Evolutionary Algorithm (PEA), and a method based on PEA for visual tracking of human body using voxel data are presented. PEA is inspired by the Quantum computation and the Quantum-inspired Evolutionary Algorithm, and it has a good balance between exploration and exploitation with very fast computation speed. The individual in PEA is encoded by the probabilistic compound bit, defined as the smallest unit of information, for the probabilistic representation. The observation step is used in PEA to obtain the observed states of the individual, and the update operator is used to evolve the individual. In the PEA based human tracking framework, tracking is considered to be a function optimization problem, so the aim is to optimize the matching function between the model and the image observation. Since the matching function is a very complex function in high-dimensional space, PEA is used to optimize it. Experiments on 3D human motion tracking using voxel data demonstrate the effectiveness, significance and computation efficiency of the proposed human tracking method.

It is well known that there are generally two possible sets of pose parameters from one calibrated perspective view of a circle. What is the relation between these two possible sets? Where does this ambiguity arise from? In which case can this ambiguity be resolved? Considering these questions, we suggest a novel viewpoint toward circle pose estimation from a single view. Different from existing methods on the basis of analytical geometry, we originally develop the projective equation of a circle, based on which a closed form solution is developed and a brand new geometric explanation for the ambiguity of solutions is presented. Experimental results verify the correctness of our proposed method.

A new calibration algorithm for multi-camera systems using the 1D calibration objects is proposed. The algorithm integrates the rank-4 factorization with Zhang’s method. The intrinsic parameters as well as the extrinsic parameters are recovered only by capturing with cameras the 1D object’s rotations around a fixed point. The algorithm is based on factorization of the scaled measurement matrix, the projective depths of which is estimated in an analytic equation instead of a recursive form. For the conditions that there are more than 3 points on the 1D object, our algorithm may solve them by extending the scaled measurement matrix. The obtained parameters are finally refined through the maximum likelihood inference. The validity of the proposed technique was verified through simulation and experiments with real images.

Grouping video contents into semantic segments is the crucial pass to content-based video summarization and retrieval. In this paper, we present a novel scene segmentation and semantic representation scheme for various video types. We first detect video shot using a coarse-to-fine algorithm. The key frames without useful information are detected and removed using template matching. Spatio-temporal coherent shots are then grouped into the same scene based on the temporal constraint of video content and visual similarity of shot activity. With general editing technique used in the continuously recorded video, semantic representation of scene content is specified to satisfy human demand on video retrieval. The proposed algorithm has been performed on various types of videos containing movie and TV program. Promising experimental results shows that the proposed method makes sense to efficient retrieval of video contents of interest.

We present a method for tracking deformable surfaces in 3D using a stereo rig. Different from traditional recursive tracking approaches that provide a strong prior on the pose for each new frame, the proposed method tracks deformable surfaces by detecting them in individual frames. In our method, the model of the surface is represented by a triangulated mesh. The constraints for model to image keypoint correspondences, together with the constraints that preserve the lengths of mesh edges, are formulated as Second Order Cone Programming (SOCP) constraints, leading this tracking-by-detection method to be an SOCP problem that can be effectively solved. Experiments on a piece of deformed paper demonstrate the capability of the proposed tracking-by-detection method.

In this paper, we present an efficient discriminative method for human pose estimation. This method learns a direct mapping from visual observations to human body configurations. The framework requires that the visual features should be powerful enough to discriminate the subtle differences between similar human poses. We propose to describe the image features using salient interest points that are represented by SIFT-like descriptors. The descriptor encode the position, appearance, and local structural information simultaneously. Bag-of-words representation is used to model the distribution of feature space. The descriptor can tolerate a range of illumination and position variations because it is computed on overlapped patches. We use Gaussian process regression to model the mapping from visual observations to human poses. This probabilistic regression algorithm is effective and robust to the pose estimation problem. We test our approach on the HumanEva data set. Experimental results demonstrate that our approach achieves the state of the art performance.

In this article, we present the projective equation of a circle in a perspective view, which naturally encodes such important geometric entities as the projected circle center, the vanishing point of the normal direction of the circle’s supporting plane and the degenerate conic envelope spanned by the image of circular points (ICPs). Based on this projective equation, we propose an easy technique to calibrate the focal length and the extrinsic parameters of a camera merely by using one perspective view of two arbitrary coplanar circles. Unlike existing optimization algorithm, our method offers a closed form solution through simple matrix manipulation. Experimental results verify the correctness and efficiency of our proposed technique.

Automatic image annotation is a promising way to achieve more effective image management and retrieval by using keywords. However, system performances of the existing state-of-the-art keyword annotation schemes are often not so satisfactory. Therefore, image annotation refinement is crucial to improve the imprecise annotation results. In this paper, a novel approach is developed to automatically refine the initial annotation of images. First, for a query image, the candidate annotations are obtained by a step-up model-based algorithm using perceptual visual characteristic. Then, a refine algorithm, fast random walk with restart is used to re-rank the candidate annotations and the top ones are reserved as the final annotations. Experiments conducted on the typical Corel dataset shows that the proposed scheme can effectively improve the automatic annotation performance.

In this letter, we present a new approach to synthesize video from still images. The method first models 3-D flows, and then it proposes image modulation. Finally, the 3-D flow models are projected onto image planes to produce the final video. We use some results of computational flow dynamics to model the flows and propose image matting and inpainting methods that are more appropriate to our animation. Our algorithm can effectively handle still images that have flows. Experiments demonstrate that our synthesis method can produce good-looking videos.

In this letter, a novel framework to segment video scene and represent scene content is proposed. Firstly, video shots are detected using a rough-to-fine algorithm. Secondly, key frames are selected adaptively, and redundant key frames are removed using template matching. Then, spatio-temporal coherent shots are clustered into the same scene. Finally, under the full analysis of typical characters on continuously recorded videos, video scene content is semantically represented to satisfy human demand on video retrieval. Experimental results show the proposed method makes sense to efficient retrieval of video content of interest.

Plane-based (2-D) camera calibration is becoming a hot research topic in recent years because of its flexibility. However, at least four image points are needed in every view to denote the coplanar feature in the 2-D camera calibration. Can we do the camera calibration by using the calibration object that only has three points? Some 1-D camera calibration techniques use the setup of three collinear points with known distances, but it is a kind of special conditions of calibration object setup. How about the general setup-three noncollinear points? We propose a new camera calibration algorithm based on the calibration objects with three noncollinear points. Experiments with simulated data and real images are carried out to verify the theoretical correctness and numerical robustness of our results. Because the objects with three noncollinear points have special properties in camera calibration, they are midway between 1-D and 2-D calibration objects. Our method is actually a new kind of camera calibration algorithm.

Vehicle Tracking is an effective approach to urban traffic flow estimation. Based on the data of GIS and GPS, a Vehicle Tracking process continually judges, from a road network, the most reliable path between every two successively detected locations of a certain vehicle, and then links the paths to restore the real track. Conventional works focused mostly on the driving distance and selected the shortest path as the solution. However, as is in many situations, the shortest path is not necessarily the best choice. Instead, a path with the least time expense may be more reasonable. Therefore, a new hybrid real-time Vehicle Tracking method is proposed in this paper considering comprehensively two main time-costing factors, the total driving distance and the number of traversed crossroads, using the concept of Pareto Frontier and Fuzzy Theory. Analytical experiment in Shanghai yields promising results.

In this study, we present an efficient approach to recover 3D human motion from monocular image sequences in generative reconstruction framework. This approach is based on the extracting of motion base space. From the motion capture data with bothersome high dimension characteristic of human activity, we extract the motion base space in which human pose can be described essentially and concisely by a more controllable way. And then, the structure of this space corresponding to some special activities such as walking motion is explored with data clustering. For the single image, Gaussian mixture model is used to generate the candidates of 3D pose. The shape context is the common descriptor of image silhouette feature and synthetical feature of human model. We get the shortlist of 3D poses by measuring the shape contexts matching cost between image feature and the synthetical features. In tracking situation, an AR model trained by the example sequences produces almost accurate pose predictions. Experiments demonstrate that the proposed approach works well

Lens distortion is one of the main factors affecting camera calibration. In this paper, a new model of camera lens distortion is presented, according to which lens distortion is governed by the coefficients of radial distortion and a transform from ideal image plane to real sensing array plane. The transform is determined by two angular parameters describing the pose and two linear parameters locating the position of the sensing array plane. Compared with the conventional ones, the new model has fewer parameters to be calibrated and more explicit physical meaning. Calibration method of the new model is also proposed. Experiments show that calibration results from the new model and method can correct lens distortion better than conventional ones

This paper presents a new method of motion analysis of articulated objects from feature point correspondences over monocular perspective images without imposing any constraints on motion. An articulated object is modeled as a kinematic chain consisting of joints and links, and its 3D joint positions are estimated within a scale factor using the connection relationship of two links over two or three images. Then, twists and exponential maps are employed to represent the motion of each link, including the general motion of the base link and the rotation of other links around their joints. Finally, constraints from image point correspondences, which are similar to that of the essential matrix in rigid motion, are developed to estimate the motion. In the algorithm, the characteristic of articulated motion, i.e., motion correlation among links, is applied to decrease the complexity of the problem and improve the robustness. A point pattern matching algorithm for articulated objects is also discussed in this paper. Simulations and experiments on real images show the correctness and efficiency of the algorithms.

We present an algebraic approach to multibody motion segmentation from line correspondences. Given three perspective views containing multiple linearly moving objects, we demonstrate that after applying a polynomial embedding to the line correspondences, they became related by the so-called multibody line constraint of translational motions. We show how to linearly estimate the multibody trifocal epipole from line-line-line correspondences. The individual trifocal epipoles are then obtained from the derivatives of the multibody line constraint (up to an unknown factor). Given normalized trifocal epipoles, we can use any special clustering technique to obtain the clustering of the motions and the correspondences. The limitations of the proposed algorithm are also discussed. Experimental results on synthetic and real dynamic scenes are presented.

This paper presents a new method of motion analysis of articulated objects from feature point correspondences over monocular perspective images without imposing any constraints on motion. The 3D joint positions of an articulated object are estimated within a scale factor using the connection relationship of two links over two or three images. Then, twists and exponential maps are employed to represent the motion of each link. Finally, constraints from image point correspondences are developed to estimate the motion. In the algorithm, the characteristic of articulated motion, i.e., motion correlation among links, is applied to decrease the complexity of the problem and improve the robustness. A point pattern matching algorithm for articulated objects is also discussed in this paper. Simulations and experiments on real images show the correctness and efficiency of the algorithms.

A model for vehicle-type motion, which assumes that the motion is a rotation around an axis through the vehicle center followed by a forward translation along the main axis of the vehicle, is proposed. When the rotation and the amplitude of translation are constant, this type of motion is shown to be equivalent to a constant camera-centered motion. This indicates that a constant motion in the conventional camera-centered model, which is commonly considered to be artificial, can in fact be a reasonable model in real life. It is shown that a constant vehicle-type motion can be interpreted as a constant screw motion. A linear algorithm for estimating constant vehicle-type motion is presented, and experiments using real scene images are included. As an extension, vehicle-type motion with constant rotation and constantly accelerated translation is discussed.>

A method for the determination of camera location from two-dimensional (2-D) to three-dimensional (3-D) straight line or point correspondences is presented. With this method, the computations of the rotation matrix and the translation vector of the camera are separable. First, the rotation matrix is found by a linear algorithm using eight or more line correspondences, or by a nonlinear algorithm using three or more line correspondences, where the line correspondences are either given or derived from point correspondences. Then, the translation vector is obtained by solving a set of linear equations based on three or more line correspondences, or two or more point correspondences. Eight 2-D to 3-D line correspondences or six 2-D to 3-D point correspondences are needed for the linear approach; three 2-D to 3-D line or point correspondences for the nonlinear approach. Good results can be obtained in the presence of noise if more than the minimum required number of correspondences are used.>

A novel method for the determination of camera location from 2-D to 3-D line or point correspondences is presented. Using this method, the computation of the rotation matrix and the translation vector of the camera are separable. First, the rotation matrix is found by a linear algorithm using eight or more line correspondences, or by a nonlinear algorithm using three of more line correspondences, where the line correspondences are either given or derived from point correspondences. Then, the translation vector can be obtained by solving a set of linear equations based on three or more line correspondences, or two or more point correspondences. Eight 2-D-to-3-D line or point correspondences or six 2-D-to-3-D point correspondences are needed for the linear approach; three 2-D-to-3-D line or point correspondences for the nonlinear approach. Good results can be obtained in the presence of noise if more than the minimum required number of correspondences are used.>

A closed-form solution to motion and structure from line correspondences in monocular perspective image sequences is presented. The algorithm requires a minimum of 13 lines over three perspective views. Redundancy in the data provides overdetermination to combat noise. The estimates can be used as an initial guess for further optimization. A unique solution to motion and structure is guaranteed if and only if the line configuration is not degenerate and the translation between any two views does not vanish. Necessary and sufficient conditions for degenerate spatial line configurations have been derived. Simulations are performed which show the performance of the algorithm in the presence of noise.>