Centre Inria de l'Université de Rennes

We address the problem of image search on a very large scale, where three constraints have to be considered jointly: the accuracy of the search, its efficiency, and the memory usage of the representation. We first propose a simple yet efficient way of aggregating local image descriptors into a vector of limited dimension, which can be viewed as a simplification of the Fisher kernel representation. We then show how to jointly optimize the dimension reduction and the indexing algorithm, so that it best preserves the quality of vector comparison. The evaluation shows that our approach significantly outperforms the state of the art: the search accuracy is comparable to the bag-of-features approach for an image representation that fits in 20 bytes. Searching a 10 million image dataset takes about 50ms.

This paper exploits the context of natural dynamic scenes for human action recognition in video. Human actions are frequently constrained by the purpose and the physical properties of scenes and demonstrate high correlation with particular scene classes. For example, eating often happens in a kitchen while running is more common outdoors. The contribution of this paper is three-fold: (a) we automatically discover relevant scene classes and their correlation with human actions, (b) we show how to learn selected scene classes from video without manual supervision and (c) we develop a joint framework for action and scene recognition and demonstrate improved recognition of both in natural video. We use movie scripts as a means of automatic supervision for training. For selected action classes we identify correlated scene classes in text and then retrieve video samples of actions and scenes for training using script-to-video alignment. Our visual models for scenes and actions are formulated within the bag-of-features framework and are combined in a joint scene-action SVM-based classifier. We report experimental results and validate the method on a new large dataset with twelve action classes and ten scene classes acquired from 69 movies.

The Critical Assessment of Metagenome Interpretation (CAMI) community initiative presents results from its first challenge, a rigorous benchmarking of software for metagenome assembly, binning and taxonomic profiling. Methods for assembly, taxonomic profiling and binning are key to interpreting metagenome data, but a lack of consensus about benchmarking complicates performance assessment. The Critical Assessment of Metagenome Interpretation (CAMI) challenge has engaged the global developer community to benchmark their programs on highly complex and realistic data sets, generated from ∼700 newly sequenced microorganisms and ∼600 novel viruses and plasmids and representing common experimental setups. Assembly and genome binning programs performed well for species represented by individual genomes but were substantially affected by the presence of related strains. Taxonomic profiling and binning programs were proficient at high taxonomic ranks, with a notable performance decrease below family level. Parameter settings markedly affected performance, underscoring their importance for program reproducibility. The CAMI results highlight current challenges but also provide a roadmap for software selection to answer specific research questions.

The first community competition designed to objectively compare the performance of particle tracking algorithms provides valuable practical information for both users and developers. Particle tracking is of key importance for quantitative analysis of intracellular dynamic processes from time-lapse microscopy image data. Because manually detecting and following large numbers of individual particles is not feasible, automated computational methods have been developed for these tasks by many groups. Aiming to perform an objective comparison of methods, we gathered the community and organized an open competition in which participating teams applied their own methods independently to a commonly defined data set including diverse scenarios. Performance was assessed using commonly defined measures. Although no single method performed best across all scenarios, the results revealed clear differences between the various approaches, leading to notable practical conclusions for users and developers.

This paper describes the OpenViBE software platform which enables researchers to design, test, and use brain–computer interfaces (BCIs). BCIs are communication systems that enable users to send commands to computers solely by means of brain activity. BCIs are gaining interest among the virtual reality (VR) community since they have appeared as promising interaction devices for virtual environments (VEs). The key features of the platform are (1) high modularity, (2) embedded tools for visualization and feedback based on VR and 3D displays, (3) BCI design made available to non-programmers thanks to visual programming, and (4) various tools offered to the different types of users. The platform features are illustrated in this paper with two entertaining VR applications based on a BCI. In the first one, users can move a virtual ball by imagining hand movements, while in the second one, they can control a virtual spaceship using real or imagined foot movements. Online experiments with these applications together with the evaluation of the platform computational performances showed its suitability for the design of VR applications controlled with a BCI. OpenViBE is a free software distributed under an open-source license.

Code and design smells are poor solutions to recurring implementation and design problems. They may hinder the evolution of a system by making it hard for software engineers to carry out changes. We propose three contributions to the research field related to code and design smells: (1) DECOR, a method that embodies and defines all the steps necessary for the specification and detection of code and design smells, (2) DETEX, a detection technique that instantiates this method, and (3) an empirical validation in terms of precision and recall of DETEX. The originality of DETEX stems from the ability for software engineers to specify smells at a high level of abstraction using a consistent vocabulary and domain-specific language for automatically generating detection algorithms. Using DETEX, we specify four well-known design smells: the antipatterns Blob, Functional Decomposition, Spaghetti Code, and Swiss Army Knife, and their 15 underlying code smells, and we automatically generate their detection algorithms. We apply and validate the detection algorithms in terms of precision and recall on XERCES v2.7.0, and discuss the precision of these algorithms on 11 open-source systems.

In the last decade, we have witnessed a drastic change in the form factor of audio and vision technologies, from heavy and grounded machines to lightweight devices that naturally fit our bodies. However, only recently, haptic systems have started to be designed with wearability in mind. The wearability of haptic systems enables novel forms of communication, cooperation, and integration between humans and machines. Wearable haptic interfaces are capable of communicating with the human wearers during their interaction with the environment they share, in a natural and yet private way. This paper presents a taxonomy and review of wearable haptic systems for the fingertip and the hand, focusing on those systems directly addressing wearability challenges. The paper also discusses the main technological and design challenges for the development of wearable haptic interfaces, and it reports on the future perspectives of the field. Finally, the paper includes two tables summarizing the characteristics and features of the most representative wearable haptic systems for the fingertip and the hand.

BACKGROUND: The process of generating raw genome sequence data continues to become cheaper, faster, and more accurate. However, assembly of such data into high-quality, finished genome sequences remains challenging. Many genome assembly tools are available, but they differ greatly in terms of their performance (speed, scalability, hardware requirements, acceptance of newer read technologies) and in their final output (composition of assembled sequence). More importantly, it remains largely unclear how to best assess the quality of assembled genome sequences. The Assemblathon competitions are intended to assess current state-of-the-art methods in genome assembly. RESULTS: In Assemblathon 2, we provided a variety of sequence data to be assembled for three vertebrate species (a bird, a fish, and snake). This resulted in a total of 43 submitted assemblies from 21 participating teams. We evaluated these assemblies using a combination of optical map data, Fosmid sequences, and several statistical methods. From over 100 different metrics, we chose ten key measures by which to assess the overall quality of the assemblies. CONCLUSIONS: Many current genome assemblers produced useful assemblies, containing a significant representation of their genes and overall genome structure. However, the high degree of variability between the entries suggests that there is still much room for improvement in the field of genome assembly and that approaches which work well in assembling the genome of one species may not necessarily work well for another.

Augmented reality (AR) allows to seamlessly insert virtual objects in an image sequence. In order to accomplish this goal, it is important that synthetic elements are rendered and aligned in the scene in an accurate and visually acceptable way. The solution of this problem can be related to a pose estimation or, equivalently, a camera localization process. This paper aims at presenting a brief but almost self-contented introduction to the most important approaches dedicated to vision-based camera localization along with a survey of several extension proposed in the recent years. For most of the presented approaches, we also provide links to code of short examples. This should allow readers to easily bridge the gap between theoretical aspects and practical implementations.

Low-cost short read sequencing technology has revolutionized genomics, though it is only just becoming practical for the high-quality de novo assembly of a novel large genome. We describe the Assemblathon 1 competition, which aimed to comprehensively assess the state of the art in de novo assembly methods when applied to current sequencing technologies. In a collaborative effort, teams were asked to assemble a simulated Illumina HiSeq data set of an unknown, simulated diploid genome. A total of 41 assemblies from 17 different groups were received. Novel haplotype aware assessments of coverage, contiguity, structure, base calling, and copy number were made. We establish that within this benchmark: (1) It is possible to assemble the genome to a high level of coverage and accuracy, and that (2) large differences exist between the assemblies, suggesting room for further improvements in current methods. The simulated benchmark, including the correct answer, the assemblies, and the code that was used to evaluate the assemblies is now public and freely available from http://www.assemblathon.org/.

Gossip-based communication protocols are appealing in large-scale distributed applications such as information dissemination, aggregation, and overlay topology management. This paper factors out a fundamental mechanism at the heart of all these protocols: the peer-sampling service. In short, this service provides every node with peers to gossip with. We promote this service to the level of a first-class abstraction of a large-scale distributed system, similar to a name service being a first-class abstraction of a local-area system. We present a generic framework to implement a peer-sampling service in a decentralized manner by constructing and maintaining dynamic unstructured overlays through gossiping membership information itself. Our framework generalizes existing approaches and makes it easy to discover new ones. We use this framework to empirically explore and compare several implementations of the peer-sampling service. Through extensive simulation experiments we show that---although all protocols provide a good quality uniform random stream of peers to each node locally---traditional theoretical assumptions about the randomness of the unstructured overlays as a whole do not hold in any of the instances. We also show that different design decisions result in severe differences from the point of view of two crucial aspects: load balancing and fault tolerance. Our simulations are validated by means of a wide-area implementation.

EMPIRE10 (Evaluation of Methods for Pulmonary Image REgistration 2010) is a public platform for fair and meaningful comparison of registration algorithms which are applied to a database of intrapatient thoracic CT image pairs. Evaluation of nonrigid registration techniques is a nontrivial task. This is compounded by the fact that researchers typically test only on their own data, which varies widely. For this reason, reliable assessment and comparison of different registration algorithms has been virtually impossible in the past. In this work we present the results of the launch phase of EMPIRE10, which comprised the comprehensive evaluation and comparison of 20 individual algorithms from leading academic and industrial research groups. All algorithms are applied to the same set of 30 thoracic CT pairs. Algorithm settings and parameters are chosen by researchers expert in the configuration of their own method and the evaluation is independent, using the same criteria for all participants. All results are published on the EMPIRE10 website (http://empire10.isi.uu.nl). The challenge remains ongoing and open to new participants. Full results from 24 algorithms have been published at the time of writing. This paper details the organization of the challenge, the data and evaluation methods and the outcome of the initial launch with 20 algorithms. The gain in knowledge and future work are discussed.

This paper addresses the modeling of reverberant recording environments in the context of under-determined convolutive blind source separation. We model the contribution of each source to all mixture channels in the time-frequency domain as a zero-mean Gaussian random variable whose covariance encodes the spatial characteristics of the source. We then consider four specific covariance models, including a full-rank unconstrained model. We derive a family of iterative expectation-maximization (EM) algorithms to estimate the parameters of each model and propose suitable procedures adapted from the state-of-the-art to initialize the parameters and to align the order of the estimated sources across all frequency bins. Experimental results over reverberant synthetic mixtures and live recordings of speech data show the effectiveness of the proposed approach.

Several recent works on action recognition have attested the importance of explicitly integrating motion characteristics in the video description. This paper establishes that adequately decomposing visual motion into dominant and residual motions, both in the extraction of the space-time trajectories and for the computation of descriptors, significantly improves action recognition algorithms. Then, we design a new motion descriptor, the DCS descriptor, based on differential motion scalar quantities, divergence, curl and shear features. It captures additional information on the local motion patterns enhancing results. Finally, applying the recent VLAD coding technique proposed in image retrieval provides a substantial improvement for action recognition. Our three contributions are complementary and lead to outperform all reported results by a significant margin on three challenging datasets, namely Hollywood 2, HMDB51 and Olympic Sports.

Standard quadrotor unmanned aerial vehicles (UAVs) possess a limited mobility because of their inherent underactuation, that is, availability of four independent control inputs (the four propeller spinning velocities) versus the 6 degrees of freedom parameterizing the quadrotor position/orientation in space. Thus, the quadrotor pose cannot track arbitrary trajectories in space (e.g., it can hover on the spot only when horizontal). Because UAVs are more and more employed as service robots for interaction with the environment, this loss of mobility due to their underactuation can constitute a limiting factor. In this paper, we present a novel design for a quadrotor UAV with tilting propellers which is able to overcome these limitations. Indeed, the additional set of four control inputs actuating the propeller tilting angles is shown to yield full actuation to the quadrotor position/orientation in space, thus allowing it to behave as a fully actuated flying vehicle. We then develop a comprehensive modeling and control framework for the proposed quadrotor, and subsequently illustrate the hardware and software specifications of an experimental prototype. Finally, the results of several simulations and real experiments are reported to illustrate the capabilities of the proposed novel UAV design.

This paper addresses recognition of human actions under view changes. We explore self-similarities of action sequences over time and observe the striking stability of such measures across views. Building upon this key observation, we develop an action descriptor that captures the structure of temporal similarities and dissimilarities within an action sequence. Despite this temporal self-similarity descriptor not being strictly view-invariant, we provide intuition and experimental validation demonstrating its high stability under view changes. Self-similarity descriptors are also shown to be stable under performance variations within a class of actions when individual speed fluctuations are ignored. If required, such fluctuations between two different instances of the same action class can be explicitly recovered with dynamic time warping, as will be demonstrated, to achieve cross-view action synchronization. More central to the current work, temporal ordering of local self-similarity descriptors can simply be ignored within a bag-of-features type of approach. Sufficient action discrimination is still retained in this way to build a view-independent action recognition system. Interestingly, self-similarities computed from different image features possess similar properties and can be used in a complementary fashion. Our method is simple and requires neither structure recovery nor multiview correspondence estimation. Instead, it relies on weak geometric properties and combines them with machine learning for efficient cross-view action recognition. The method is validated on three public data sets. It has similar or superior performance compared to related methods and it performs well even in extreme conditions, such as when recognizing actions from top views while using side views only for training.

Emergence of polyphagous herbivorous insects entails significant adaptation to recognize, detoxify and digest a variety of host-plants. Despite of its biological and practical importance - since insects eat 20% of crops - no exhaustive analysis of gene repertoires required for adaptations in generalist insect herbivores has previously been performed. The noctuid moth Spodoptera frugiperda ranks as one of the world's worst agricultural pests. This insect is polyphagous while the majority of other lepidopteran herbivores are specialist. It consists of two morphologically indistinguishable strains ("C" and "R") that have different host plant ranges. To describe the evolutionary mechanisms that both enable the emergence of polyphagous herbivory and lead to the shift in the host preference, we analyzed whole genome sequences from laboratory and natural populations of both strains. We observed huge expansions of genes associated with chemosensation and detoxification compared with specialist Lepidoptera. These expansions are largely due to tandem duplication, a possible adaptation mechanism enabling polyphagy. Individuals from natural C and R populations show significant genomic differentiation. We found signatures of positive selection in genes involved in chemoreception, detoxification and digestion, and copy number variation in the two latter gene families, suggesting an adaptive role for structural variation.

How do people appropriate their virtual hand representation when interacting in virtual environments? In order to answer this question, we conducted an experiment studying the sense of embodiment when interacting with three different virtual hand representations, each one providing a different degree of visual realism but keeping the same control mechanism. The main experimental task was a Pick-and-Place task in which participants had to grasp a virtual cube and place it to an indicated position while avoiding an obstacle (brick, barbed wire or fire). An additional task was considered in which participants had to perform a potentially dangerous operation towards their virtual hand: place their virtual hand close to a virtual spinning saw. Both qualitative measures and questionnaire data were gathered in order to assess the sense of agency and ownership towards each virtual hand. Results show that the sense of agency is stronger for less realistic virtual hands which also provide less mismatch between the participant's actions and the animation of the virtual hand. In contrast, the sense of ownership is increased for the human virtual hand which provides a direct mapping between the degrees of freedom of the real and virtual hand.

With increasing numbers of energy hungry data centers energy conservation has now become a major design constraint. One traditional approach to conserve energy in virtualized data centers is to perform workload (i.e., VM) consolidation. Thereby, workload is packed on the least number of physical machines and over-provisioned resources are transitioned into a lower power state. However, most of the workload consolidation approaches applied until now are limited to a single resource (e.g., CPU) and rely on simple greedy algorithms such as First-Fit Decreasing (FFD), which perform resource-dissipative workload placement. Moreover, they are highly centralized and known to be hard to distribute. In this work, we model the workload consolidation problem as an instance of the multi-dimensional bin-packing (MDBP) problem and design a novel, nature-inspired workload consolidation algorithm based on the Ant Colony Optimization (ACO). We evaluate the ACO-based approach by comparing it with one frequently applied greedy algorithm (i.e., FFD). Our simulation results demonstrate that ACO outperforms the evaluated greedy algorithm as it achieves superior energy gains through better server utilization and requires less machines. Moreover, it computes solutions which are nearly optimal. Finally, the autonomous nature of the approach allows it to be implemented in a fully distributed environment.