Automatic quantification of the left ventricle (LV) from cardiac magnetic resonance (CMR) images plays an important role in making the diagnosis procedure efficient, reliable, and alleviating the laborious reading work for physicians. Considerable efforts have been devoted to LV quantification using different strategies that include segmentation-based (SG) methods and the recent direct regression (DR) methods. Although both SG and DR methods have obtained great success for the task, a systematic platform to benchmark them remains absent because of differences in label information during model learning. In this paper, we conducted an unbiased evaluation and comparison of cardiac LV quantification methods that were submitted to the Left Ventricle Quantification (LVQuan) challenge, which was held in conjunction with the Statistical Atlases and Computational Modeling of the Heart (STACOM) workshop at the MICCAI 2018. The challenge was targeted at the quantification of 1) areas of LV cavity and myocardium, 2) dimensions of the LV cavity, 3) regional wall thicknesses (RWT), and 4) the cardiac phase, from mid-ventricle short-axis CMR images. First, we constructed a public quantification dataset Cardiac-DIG with ground truth labels for both the myocardium mask and these quantification targets across the entire cardiac cycle. Then, the key techniques employed by each submission were described. Next, quantitative validation of these submissions were conducted with the constructed dataset. The evaluation results revealed that both SG and DR methods can offer good LV quantification performance, even though DR methods do not require densely labeled masks for supervision. Among the 12 submissions, the DR method LDAMT offered the best performance, with a mean estimation error of 301 mm2for the two areas, 2.15 mm for the cavity dimensions, 2.03 mm for RWTs, and a 9.5% error rate for the cardiac phase classification. Three of the SG methods also delivered comparable performances. Finally, we discussed the advantages and disadvantages of SG and DR methods, as well as the unsolved problems in automatic cardiac quantification for clinical practice applications.

Delineation of the left ventricular cavity, myocardium, and right ventricle from cardiac magnetic resonance images (multi-slice 2-D cine MRI) is a common clinical task to establish diagnosis. The automation of the corresponding tasks has thus been the subject of intense research over the past decades. In this paper, we introduce the "Automatic Cardiac Diagnosis Challenge" dataset (ACDC), the largest publicly available and fully annotated dataset for the purpose of cardiac MRI (CMR) assessment. The dataset contains data from 150 multi-equipments CMRI recordings with reference measurements and classification from two medical experts. The overarching objective of this paper is to measure how far state-of-the-art deep learning methods can go at assessing CMRI, i.e., segmenting the myocardium and the two ventricles as well as classifying pathologies. In the wake of the 2017 MICCAI-ACDC challenge, we report results from deep learning methods provided by nine research groups for the segmentation task and four groups for the classification task. Results show that the best methods faithfully reproduce the expert analysis, leading to a mean value of 0.97 correlation score for the automatic extraction of clinical indices and an accuracy of 0.96 for automatic diagnosis. These results clearly open the door to highly accurate and fully automatic analysis of cardiac CMRI. We also identify scenarios for which deep learning methods are still failing. Both the dataset and detailed results are publicly available online, while the platform will remain open for new submissions.

In this paper, we propose an efficient clustering algorithm that has been applied to the microaggregation problem. The goal is to partition N given records into clusters, each of them grouping at least K records, so that the sum of the within-partition squared error (SSE) is minimized. We propose a successive Group Selection algorithm that approximately solves the microaggregation problem in O(N2log N) time, based on sequential Minimization of SSE. Experimental results and comparisons to existing methods with similar computation cost on real and synthetic data sets demonstrate the high performance and robustness of the proposed scheme.

This paper introduces a new rigorous theoretical framework to address discrete MRF-based optimization in computer vision. Such a framework exploits the powerful technique of Dual Decomposition. It is based on a projected subgradient scheme that attempts to solve an MRF optimization problem by first decomposing it into a set of appropriately chosen subproblems, and then combining their solutions in a principled way. In order to determine the limits of this method, we analyze the conditions that these subproblems have to satisfy and demonstrate the extreme generality and flexibility of such an approach. We thus show that by appropriately choosing what subproblems to use, one can design novel and very powerful MRF optimization algorithms. For instance, in this manner we are able to derive algorithms that: 1) generalize and extend state-of-the-art message-passing methods, 2) optimize very tight LP-relaxations to MRF optimization, and 3) take full advantage of the special structure that may exist in particular MRFs, allowing the use of efficient inference techniques such as, e.g., graph-cut-based methods. Theoretical analysis on the bounds related with the different algorithms derived from our framework and experimental results/comparisons using synthetic and real data for a variety of tasks in computer vision demonstrate the extreme potentials of our approach.

We propose a general purpose image segmentation framework, which involves feature extraction and classification in feature space, followed by flooding and merging in spatial domain. Region growing is based on the computed local measurements and distances from the distribution of features describing the different classes. Using the properties of the label dependent distances spatial coherence is ensured, since the image features are described globally. The distribution of the features for the different classes are obtained by block-wise unsupervised clustering based on the construction of the minimum spanning tree of the blocks' grid using the Mallows distance and the equipartition of the resulting tree. The final clustering is obtained by using the k-centroids algorithm. With high probability and under topological constraints, connected components of the maximum likelihood classification map are used to compute a map of initially labelled pixels. An efficient flooding algorithm is introduced, namely, Priority Multi-Class Flooding Algorithm (PMCFA), that assign pixels to labels using Bayesian dissimilarity criteria. A new region merging method, which incorporates boundary information, is introduced for obtaining the final segmentation map. Therefore, the merging stage is based on region features and edge localization. Segmentation results on the Berkeley benchmark data set demonstrate the effectiveness of the proposed methods.

We propose a method for interactive colour image segmentation. The goal is to detect an object from the background, when some markers on object(s) and the background are given. As features only probability distributions of the data are used. At first, all the labelled seeds are independently propagated for obtaining homogeneous connected components for each of them. Then the image is divided in blocks, which are classified according to their probabilistic distance from the classified regions. A topographic surface for each class is obtained, using Bayesian dissimilarities and a min-max criterion. Two algorithms are proposed: a regularized classification based on the topographic surface and incorporating an MRF model, and a priority multi-label flooding algorithm. Segmentation results on the LHI data set are presented.

In this paper we introduce a novel approach to single view reconstruction using shape grammars. Our approach consists in modeling architectural styles using a set of basic shapes and a set of parametric rules, corresponding to increasing levels of detail. This approach is able to model elaborate and varying architectural styles, using a tree representation of variable depth and complexity. Towards reconstruction, the parameters of the rules are optimized using image-based and architectural costs. This is done through an efficient MRF formulation based on the shape grammar itself. The resulting framework can produce precise 3D models from single views, can deal with lack of texture and the presence of occlusions and specular reflections, while maintaining the ability to cope with very complex architectural styles. Promising results demonstrate the potential of our approach.

We present a key frames selection algorithm based on three iso-content principles (iso-content distance, iso-content error and iso-content distortion), so that the selected key frames are equidistant in video content according to the used principle. Two automatic approaches for defining the most appropriate number of key frames are proposed by exploiting supervised and unsupervised content criteria. Experimental results and the comparisons with existing methods from literature on large dataset of real-life video sequences illustrate the high performance of the proposed schemata.

In this paper, we propose a novel dynamic discrete framework to address image morphing with application to optical flow estimation. We reformulate the problem using a number of discrete displacements, and therefore the estimation of the morphing parameters becomes a tractable matching criteria independent combinatorial problem which is solved through the FastPD algorithm. In order to overcome the main limitation of discrete approaches (low dimensionality of the label space is unable to capture the continuous nature of the expected solution), we introduce a dynamic behavior in the model where the plausible discrete deformations (displacements) are varying in space (across the domain) and time (different states of the process - successive morphing states) according to the local uncertainty of the obtained solution.

A novel center-based clustering algorithm is proposed in this paper. We first formulate clustering as an NP-hard linear integer program and we then use linear programming and the duality theory to derive the solution of this optimization problem. This leads to an efficient and very general algorithm, which works in the dual domain, and can cluster data based on an arbitrary set of distances. Despite its generality, it is independent of initialization (unlike EM-like methods such as K-means), has guaranteed convergence, and can also provide online optimality bounds about the quality of the estimated clustering solutions. To deal with the most critical issue in a center-based clustering algorithm (selection of cluster centers), we also introduce the notion of stability of a cluster center, which is a well defined LP-based quantity that plays a key role to our algorithm's success. Furthermore, we also introduce, what we call, the margins (another key ingredient in our algorithm), which can be roughly thought of as dual counterparts to stabilities and allow us to obtain computationally efficient approximations to the latter. Promising experimental results demonstrate the potentials of our method.

We present a shape-based method for automatic people detection and counting without any assumption concerning camera motion. In order to evaluate the robustness of the proposed method, we apply it for classifying athletics videos into two classes: videos of individual and videos of team sports. The videos used are real and characterized by dynamic and unconstrained environment. Moreover, in the case of team sport, we propose a shape deformations based method for running/hurdling discrimination (activity recognition). Robust, adaptive and independent from color, illumination changes and the camera motion, the proposed features are combined in the Transferable Belief Model (TBM) framework providing a two-level (frames and shot) video categorization. Experimental results of 97% of accuracy for individual/team sport categorization using a dataset of 252 real videos of athletic meetings, acquired by moving cameras under varying view angles, indicate the stability and the good performance of the proposed scheme.

A Bayesian, fully automatic moving object localization method is proposed, using inter-frame differences and background/foreground colour as discrimination cues. Change detection pixel classification to one of the labels "changed" or "unchanged" is obtained by mixture analysis, while histograms are used for statistical description of colours. High confidence, change detection based, statistical criteria are used to compute a map of initial labelled pixels. Finally, a region growing algorithm, which is named priority multi-label flooding algorithm, assigns pixels to labels using Bayesian dissimilarity criteria. Localization results on well-known benchmark image sequences as well as on webcam and compressed videos are presented.

We propose a general framework that focuses on automatic individual/multiple people motion-shape analysis and on suitable features extraction that can be used on action/activity recognition problems under real, dynamical and unconstrained environments. We have considered various athletic videos from a single uncalibrated, possibly moving camera in order to evaluate the robustness of the proposed method. We have used an easily expanded hierarchical scheme in order to classify them to videos of individual and team sports. Robust, adaptive and independent from the camera motion, the proposed features are combined within Transferable Belief Model (TBM) framework providing a two level (frames and shot) video categorization. The experimental results of 97% individual/team sport categorization accuracy, using a dataset of more than 250 videos of athletic meetings indicate the good performance of the proposed scheme.

A new efficient MRF optimization algorithm, called Fast-PD, is proposed, which generalizes α-expansion. One of its main advantages is that it offers a substantial speedup over that method, e.g. it can be at least 3-9 times faster than α-expansion. Its efficiency is a result of the fact that Fast-PD exploits information coming not only from the original MRF problem, but also from a dual problem. Furthermore, besides static MRFs, it can also be used for boosting the performance of dynamic MRFs, i.e. MRFs varying over time. On top of that, Fast-PD makes no compromise about the optimality of its solutions: it can compute exactly the same answer as a-expansion, but, unlike that method, it can also guarantee an almost optimal solution for a much wider class of NP-hard MRF problems. Results on static and dynamic MRFs demonstrate the algorithm's efficiency and power. E.g., Fast-PD has been able to compute disparity for stereoscopic sequences in real time, with the resulting disparity coinciding with that of a-expansion.

A new message-passing scheme for MRF optimization is proposed in this paper. This scheme inherits better theoretical properties than all other state-of-the-art message passing methods and in practice performs equally well/outperforms them. It is based on the very powerful technique of Dual Decomposition [1] and leads to an elegant and general framework for understanding/designing message-passing algorithms that can provide new insights into existing techniques. Promising experimental results and comparisons with the state of the art demonstrate the extreme theoretical and practical potentials of our approach.

In this paper we present a video summarization scheme. First, shot detection is performed and then we extract the key frames under an equality requirement on subshots. We propose a key frames selection algorithm (Iso-Content MINMAX), which is very flexible on any choice of content descriptors, and is based on MINMAX optimization formulation. The equality principle provides to the selected key frames the useful property to be equivalent on video content summarization.

A new framework is presented for both understanding and developing graph-cut-based combinatorial algorithms suitable for the approximate optimization of a very wide class of Markov Random Fields (MRFs) that are frequently encountered in computer vision. The proposed framework utilizes tools from the duality theory of linear programming in order to provide an alternative and more general view of state-of-the-art techniques like the \alpha-expansion algorithm, which is included merely as a special case. Moreover, contrary to \alpha-expansion, the derived algorithms generate solutions with guaranteed optimality properties for a much wider class of problems, for example, even for MRFs with nonmetric potentials. In addition, they are capable of providing per-instance suboptimality bounds in all occasions, including discrete MRFs with an arbitrary potential function. These bounds prove to be very tight in practice (that is, very close to 1), which means that the resulting solutions are almost optimal. Our algorithms' effectiveness is demonstrated by presenting experimental results on a variety of low-level vision tasks, such as stereo matching, image restoration, image completion, and optical flow estimation, as well as on synthetic problems.

In this paper, a new exemplar-based framework is presented, which treats image completion, texture synthesis, and image inpainting in a unified manner. In order to be able to avoid the occurrence of visually inconsistent results, we pose all of the above image-editing tasks in the form of a discrete global optimization problem. The objective function of this problem is always well-defined, and corresponds to the energy of a discrete Markov random field (MRF). For efficiently optimizing this MRF, a novel optimization scheme, called priority belief propagation (BP), is then proposed, which carries two very important extensions over the standard BP algorithm: "priority-based message scheduling" and "dynamic label pruning." These two extensions work in cooperation to deal with the intolerable computational cost of BP, which is caused by the huge number of labels associated with our MRF. Moreover, both of our extensions are generic, since they do not rely on the use of domain-specific prior knowledge. They can, therefore, be applied to any MRF, i.e., to a very wide class of problems in image processing and computer vision, thus managing to resolve what is currently considered as one major limitation of the BP algorithm: its inefficiency in handling MRFs with very large discrete state spaces. Experimental results on a wide variety of input images are presented, which demonstrate the effectiveness of our image-completion framework for tasks such as object removal, texture synthesis, text removal, and image inpainting.

In this paper, an adaptable neural network model is used for real time video delivery over communication networks of low and variable bandwidth, such as the wireless ones. The scheme performs video delivery in content domain in contrast to the previous approaches in which only temporal frame skipping is adopted. The proposed method requires no buffering of video frames and thus imposing no frame delay. In particular, in case of low bandwidth conditions, the proposed scheme estimates the number of frames that best represent the sequence within a time segment and transmit this number for delivery instead of a temporal frame skipping. Multiple key frames are considered by optimally approximating the real bandwidth availability with a rational fraction. Key frame estimation is accomplished using a neural network model capable of predicting the indices of the most appropriate key frames that are to be delivered without being available the video information. The model takes into account the previous information as it has been evaluated by the already delivered information. The proposed scheme is based on an efficient recursive estimation algorithm since the network weights cannot be considered constant throughout video transmission. This is due to the fact that content as well as bandwidth characteristics vary from time to time.

A new framework is presented that uses tools from duality theory of linear programming to derive graph-cut based combinatorial algorithms for approximating NP-hard classification problems. The derived algorithms include alpha-expansion graph cut techniques merely as a special case, have guaranteed optimality properties even in cases where alpha-expansion techniques fail to do so and can provide very tight per-instance suboptimality bounds in all occasions

Over the last several years, major efforts have been made to develop methods for extracting information from audiovisual media, in order that they may be stored and retrieved in databases automatically, based on their content. In this work we deal with the characterization of an audio signal, which may be part of a larger audiovisual system or may be autonomous, as for example in the case of an audio recording stored digitally on disk. Our goal was to first develop a system for segmentation of the audio signal, and then classification into one of two main categories: speech or music. Among the system's requirements are its processing speed and its ability to function in a real-time environment with a small responding delay. Because of the restriction to two classes, the characteristics that are extracted are considerably reduced and moreover the required computations are straightforward. Experimental results show that efficiency is exceptionally good, without sacrificing performance. Segmentation is based on mean signal amplitude distribution, whereas classification utilizes an additional characteristic related to the frequency. The classification algorithm may be used either in conjunction with the segmentation algorithm, in which case it verifies or refutes a music-speech or speech-music change, or autonomously, with given audio segments. The basic characteristics are computed in 20 ms intervals, resulting in the segments' limits being specified within an accuracy of 20 ms. The smallest segment length is one second. The segmentation and classification algorithms were benchmarked on a large data set, with correct segmentation about 97% of the time and correct classification about 95%.

In this paper, we explore image retrieval mechanisms based on a combination of texture and color features. Texture features are extracted using Discrete Wavelet Frames (DWF) analysis, an over-complete decomposition in scale and orientation. Two-dimensional (2-D) or one-dimensional (1-D) histograms of the CIE Lab chromaticity coordinates are used as color features. The 1-D histograms of the a, b coordinates were modeled according to the generalized Gaussian distribution. The similarity measure defined on the feature distribution is based on the Bhattacharya distance. Retrieval benchmarking is performed over the Brodatz album and on images from natural scenes, obtained from the VisTex database of MIT Media Laboratory and from the Corel Photo Gallery. As a performance indicator recall (relative number of correct images retrieved) is measured on both texture and color separately and in combination. Experiments show this approach to be as effective as other methods while computationally more tractable.

This paper introduces a level set methodology for the precise boundary localization of image objects within an indicated region, designed to be particularly robust against weak or spurious edges, triple points or inhomogeneity of object features in the proximity of the actual interface. The proposed technique requires a reliable classification for a subset of the object interiors, which is propagated towards the unclassified space using a competitive, statistically motivated fast marching region growing algorithm. Color and texture features are used on a locally adaptive, dynamically updated fashion to allow for the robust discrimination of inhomogeneous objects and an efficient implementation. Applications are illustrated in the context of moving object localization and semiautomatic object extraction.

Recovering 3D motion parameters from 2D displacements is a difficult task, given the influence of noise contained in these data, which correspond at best to a crude approximation of the real motion field. The need for stability of the system of equations to be solved is therefore essential. We present a novel method based on an unbiased estimator that aims at enhancing this stability and strongly reduces the influence of noise contamination. Experimental results using synthetic and real optical flows are presented to demonstrate the effectiveness of our method in comparison to a set of selected methods.

The segmentation of colored texture images is considered. Either luminance, color, and/or texture features could be used for segmentation. For luminance and color the classes are described using the corresponding empirical probability distributions. The discrete wavelet frames analysis is used for obtaining features of texture patterns. At a first stage, pattern analysis is performed for extracting the features using the Bhattacharya distance. Two labeling algorithms are proposed. A deterministic relaxation algorithm using a likelihood based distance yields the labeling of pixels to the different color-texture patterns. In addition, a multi-label fast marching level set algorithm is utilized for the determination of the segment boundaries.

Detecting and recognizing human faces automatically in digital images strongly enhance content-based video indexing systems. In this paper, a novel scheme for human faces detection in color images under nonconstrained scene conditions, such as the presence of a complex background and uncontrolled illumination, is presented. Color clustering and filtering using approximations of the YCbCr and HSV skin color subspaces are applied on the original image, providing quantized skin color regions. A merging stage is then iteratively performed on the set of homogeneous skin color regions in the color quantized image, in order to provide a set of potential face areas. Constraints related to shape and size of faces are applied, and face intensity texture is analyzed by performing a wavelet packet decomposition on each face area candidate in order to detect human faces. The wavelet coefficients of the band filtered images characterize the face texture and a set of simple statistical deviations is extracted in order to form compact and meaningful feature vectors. Then, an efficient and reliable probabilistic metric derived from the Bhattacharrya distance is used in order to classify the extracted feature vectors into face or nonface areas, using some prototype face area vectors, acquired in a previous training stage.

In this paper we address an important issue in motion analysis: the detection of moving objects. A statistical approach is adopted in order to formulate the problem. The inter-frame difference is modeled by a mixture of Laplacian distributions, and a Gibbs random field is used for describing the label set. A new method to determine the regularization parameter is proposed, based on a voting technique. Then two different multiscale algorithms are evaluated, and the labeling problem is solved using either ICM (iterated conditional modes) or HCF (highest confidence first) algorithms. Experimental results are provided using synthetic and real video sequences.

Two important problems in motion analysis are addressed in this paper: change detection and moving object location. For the first problem, the inter-frame difference is modelized by a mixture of Laplacian distributions, a Gibbs random field is used for describing the label field, and HCF (highest confidence first) algorithm is used for solving the resulting optimization problem. The solution of the second problem is based on the observation of two successive frames alone. Using the results of change detection an adaptive statistical model for the couple of image intensities is identified. Then the labeling problem is solved using HCF algorithm. Results on real image sequences illustrate the efficiency of the proposed method.

This work aims at determining, given two stereoscopic image sequences, at any time instant two dense velocity fields, for the left and the right sequence, and the disparity field. The disparity field of the previous stereoscopic pair is considered as known. Thus at the initial time instant the disparity field of the first stereoscopic pair is estimated. For both problems multiscale iterative relaxation algorithms are used. Results are given with real stereoscopic data.

The robustness of adaptive differential pulse code modulation (ADPCM) systems versus transmission errors is addressed. To secure the stability of the decoder, it is necessary to modify the form of the LMS (least-mean-square) algorithm used to adapt the predictor. Solutions introducing soft constraints are investigated. The leakage algorithm is proved to be not fully satisfactory, and a new stabilizing algorithm is presented that makes it possible to achieve good performances. Compared to existing methods, the main advantage of this algorithm is its low computational complexity. Form a theoretical point of view, the effect of transmission errors is described by a set of nonlinear recurrent equations. An analysis is carries out in the deterministic second-order case.>

A hybrid image predictive coding method is presented. The intraframe predictor is an adaptive FIR (finite impulse response) filter that uses the LMS (least mean squares) algorithm to track spatial local characteristics of the intensity continuously. The interframe predictor is motion-adaptive, using a pel-recursive method to estimate the displacement vector. A weight coefficient is adapted continuously in order to favor the prediction mode that performs better. For the sequence examined a significant improvement is obtained in comparison with either mode alone. A crucial problem in predictive coding, particularly with adaptive techniques, is that of sensitivity to transmission errors. A method ensuring the autoadjustment of the decoder in the presence of isolated transmission errors is proposed for the intraframe mode. Neither overhead information nor an error-correcting code are needed.>

Quadratic signal processing is used in detection and estimation of random signals. To describe the performance of quadratic signal processing, the probability distribution of the output of the processor is needed. Only positive-definite Gaussian quadratic forms are considered. The quadratic form is diagonalized in terms of independent Gaussian variables and its mean, moment-generating function, and cumulants are computed; conditions are given for the quadratic form to be\chi^{2}distributed and distributed like a sum of independent random variables having a Gamma distribution. A new method is proposed to approximate its probability distribution using an expansion in Laguerre polynomials for the central case and in generalized\chi^{2}distributions in the noncentral case. The series coefficients and bounds on truncation error are evaluated. Some applications in average power and power spectrum estimation and in detection illustrate our method.

When the transmission is fluctuating (with respect to the elementary transmission duration) diversity techniques enable to improve the performance. In a first part the Rayleigh model has been tested in long distance submarine acoustics transmission and we present some experimental results of explicit frequential diversity. In the general delay and Doppler dispersive case, we present some results of implicit diversity in order to decrease the error probability and we show the consequences of this concept on to the signal design problem. We have applied this study particularly in two cases : the multipath transmission and the fast-fading modulation case.