Articles

Combining 3D shape and color for 3D object recognition

Brandao S., Costeira J.P., Veloso M.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

2016

Abstract:

We present new results in object recognition based on color and 3D shape obtained from 3D cameras. Namely, we further exploit diffusion processes to represent shape and the use of color/texture as a perturbation to the diffusion process. Diffusion processes are an effective tool to replace shortest path distances in the characterization of 3D shapes. They also provide effective means for the seamlessly representation of color and shape, mainly because they provide information both the color and on their distribution over surfaces. While there have been different approaches for incorporating color information in the diffusion process, this is the first work that explores different parameterizations of color and their impact on recognition tasks. We present results using very challenging datasets, where we propose to recognize different instances of the same object class assuming a very limited a-priori knowledge on each individual object.

Combining free energy score spaces with information theoretic kernels: Application to scene classification

Bicego M., Perina A., Murino V., Martins A., Aguiar P., Figueiredo M.

Proceedings - International Conference on Image Processing, ICIP

2010

Abstract:

Most approaches to learn classifiers for structured objects (e.g., images) use generative models in a classical Bayesian framework. However, state-of-the-art classifiers for vectorial data (e.g., support vector machines) are learned discriminatively. A generative embedding is a mapping from the object space into a fixed dimensional score space, induced by a generative model, usually learned from data. The fixed dimensionality of these generative score spaces makes them adequate for discriminative learning of classifiers, thus bringing together the best of the discriminative and generative paradigms. In particular, it was recently shown that this hybrid approach outperforms a classifier obtained directly for the generative model upon which the score space was built. Using a generative embedding involves two steps: (i) defining and learning the generative model and using it to build the embedding; (ii) discriminatively learning a (maybe kernel) classifier on the adopted score space. The literature on generative embeddings is essentially focused on step (i), usually using some standard off-the-shelf tool for step (ii). In this paper, we adopt a different approach, by focusing also on the discriminative learning step. In particular, we combine two very recent and top performing tools in each of the steps: (i) the free energy score space; (ii) non-extensive information theoretic kernels. In this paper, we apply this methodology in scene recognition. Experimental results on two benchmark datasets shows that our approach yields state-of-the-art performance.

Combining general multi-class and specific two-class classifiers for improved customized ECG heartbeat classification

Ye C., Kumar B.V.K.V., Coimbra M.T.

Proceedings - International Conference on Pattern Recognition

2012

Abstract:

We present an approach for customized heartbeat classification of electrocardiogram (ECG) signals, based on the construction of one general multi-class classifier and one specific two-class classifier. The general classifier is trained on a global training dataset, containing examples of all possible classes and patterns. On the other hand, the individual-specific classifier is built using a small amount of individual data, which is a binary one-against-the-rest classifier, providing discrimination between normal and abnormal patterns from that individual. Such an individual-specific classifier can be a two-class classifier or a one-class classifier, depending on the availability of abnormal patterns in the individual training dataset. The classifications from the two classifiers are fused to obtain a final decision. The proposed approach is applied to the study of ECG heartbeat classification problem, significantly outperforming state-of-the-art methods. The proposed method can also be useful in anomaly detection of other biomedical signals.

Combining information theoretic kernels with generative embeddings for classification

Bicego M., Ulas A., Castellani U., Perina A., Murino V., Martins A.F.T., Aguiar P.M.Q., Figueiredo M.A.T.

Neurocomputing

2013

Abstract:

Classical approaches to learn classifiers for structured objects (e.g., images, sequences) use generative models in a standard Bayesian framework. To exploit the state-of-the-art performance of discriminative learning, while also taking advantage of generative models of the data, generative embeddings have been recently proposed as a way of building hybrid discriminative/generative approaches. A generative embedding is a mapping, induced by a generative model (usually learned from data), from the object space into a fixed dimensional space, adequate for discriminative classifier learning. Generative embeddings have been shown to often outperform the classifiers obtained directly from the generative models upon which they are built. Using a generative embedding for classification involves two main steps: (i) defining and learning a generative model and using it to build the embedding; (ii) discriminatively learning a (maybe kernel) classifier with the embedded data. The literature on generative embeddings is essentially focused on step (i), usually taking some standard off-the-shelf tool for step (ii). Here, we adopt a different approach, by focusing also on the discriminative learning step. In particular, we exploit the probabilistic nature of generative embeddings, by using kernels defined on probability measures; in particular we investigate the use of a recent family of non-extensive information theoretic kernels on the top of different generative embeddings. We show, in different medical applications that the approach yields state-of-the-art performance.

Comments on Jacobson et al.’s proposal for a wind, water, and solar energy future for New York State

Gilbraith N., Jaramillo P., Tong F., Faria F.

Energy Policy

2013

Abstract:

Jacobson et al. (2013) recently published a paper arguing the feasibility of meeting all of the energy demands in New York State with wind, solar, and water resources. In this forum we suggest that the authors do not present sufficient analysis to demonstrate the technical, economic, and social feasibility of their proposed strategy.

Composing interfering abstract: Protocols

Militao F., Aldrich J., Caires L.

Leibniz International Proceedings in Informatics, LIPIcs

2016

Abstract:

The undisciplined use of shared mutable state can be a source of program errors when aliases unsafely interfere with each other. While protocol-based techniques to reason about interference abound, they do not address two practical concerns: the decidability of protocol composition and its integration with protocol abstraction. We show that our composition procedure is decidable and that it ensures safe interference even when composing abstract protocols. To evaluate the expressiveness of our protocol framework for safe shared memory interference, we show how this same protocol framework can be used to model safe, typeful message-passing concurrency idioms.

Concise integer linear programming formulations for dependency parsing

Martins A.F.T., Smith N.A., Xing E.P.

ACL-IJCNLP 2009 - Joint Conf. of the 47th Annual Meeting of the Association for Computational Linguistics and 4th Int. Joint Conf. on Natural Language Processing of the AFNLP, Proceedings of the Conf.

2009

Abstract:

We formulate the problem of non-projective dependency parsing as a polynomial-sized integer linear program. Our formulation is able to handle non-local output features in an efficient manner; not only is it compatible with prior knowledge encoded as hard constraints, it can also learn soft constraints from data. In particular, our model is able to learn correlations among neighboring arcs (siblings and grandparents), word valency, and tendencies toward nearly-projective parses. The model parameters are learned in a max-margin framework by employing a linear programming relaxation. We evaluate the performance of our parser on data in several natural languages, achieving improvements over existing state-of-the-art methods.

Concurrency by default: Using permissions to express dataflow in stateful programs

Stork S., Marques P., Aldrich J.

Proceedings of the Conference on Object-Oriented Programming Systems, Languages, and Applications, OOPSLA

2009

Abstract:

The rise of the multicore era is catapulting concurrency into mainstream programming. Current programming paradigms build in sequentiality, and as a result, concurrency support in those languages forces programmers into low-level reasoning about execution order. In this paper, we introduce a new programming paradigm in which concurrency is the default. Our Aeminium language uses access permissions to express logical dependencies in the code at a higher level of abstraction than sequential order. Therefore compiler/runtime-system can leverage that dependency information to allow concurrent execution. Because in Aeminium programmers specify dependencies rather than control flow, there is no need to engage in difficult, error-prone, and low-level reasoning about execution order or thread interleavings. Developers can instead focus on the design of the program, and benefit as the runtime automatically extracts the concurrency inherent in

Consensus and products of random stochastic matrices: Exact rate for convergence in probability

Bajovic D., Xavier J., Moura J.M.F., Sinopoli B.

IEEE Transactions on Signal Processing

2013

Abstract:

We find the exact rate for convergence in probability of products of independent, identically distributed symmetric, stochastic matrices. It is well-known that if the matrices have positive diagonals almost surely and the support graph of the mean or expected value of the random matrices is connected, the products of the matrices converge almost surely to the average consensus matrix, and thus in probability. In this paper, we show that the convergence in probability is exponentially fast, and we explicitly characterize the exponential rate of this convergence. Our analysis reveals that the exponential rate of convergence in probability depends only on the statistics of the support graphs of the random matrices. Further, we show how to compute this rate for commonly used random models: gossip and link failure. With these models, the rate is found by solving a min-cut problem, and hence it is easily computable. Finally, as an illustration, we apply our results to solving power allocation among networked sensors in a consensus+innovations distributed detection problem.