Retrieving specific categories of images among billions of images usually requires an annotation step. Unfortunately, keywords-based techniques suffer from the semantic gap existing between a semantic concept and its digital representation. Content Based Image Retrieval (CBIR) systems tackle this issue simply considering semantic proximities can be mapped to similarities in the image space. Introducing relevance feedbacks involves the user in the task, but extends the annotation step.

The ability to obtain a compact description of an object and to keep the significant information can be crucial for many applications such as indexing, clustering and classification. Formal concept analysis (FCA) provides an algorithm to perform such dimension reduction. However, the requirements of FCA limit its performance. In order to relax FCA requirements, this paper presents an extension of FCA with the definition and formal analysis of a fuzzy precedence graph. This is completed by a fuzzy dimension reduction algorithm. We have tested our algorithm on several real datasets related to bag of visual words and evaluated it for a classification task. The proposed formalism and algorithm improve FCA tools on several cases while retaining the significant information. Furthermore, the fuzzy dimension reduction algorithm never performs worse than FCA tools in terms of dimension reduction. We have also compared it with PCA dimension reduction. PCA usually reduces more attributes but looses more information and does so in an unreliable/unstable manner compared to our algorithm.

This paper proposes a place cell model allowing place recognition in the context of robot autonomous navigation. The robustness of this approach lies in the fact that even if one or several patterns characterizing the place are removed or not visible anymore, a place can still be recognized. The recognition process in this work is improved with respect to the state-of-the-art place cells approach. Additionally, the interconnection of the modules is made such that the robot is able to learn new places as it navigates and interacts with the environment to get to its final destination. Experimental results validate the advantage of the incremental learning allowing the robot to cope with any unforeseen changes and thus adapting itself to the environment.

This paper proposes the analysis of a paper-based floor plan of a building for reference purposes in the navigation task of a mobile robot. The novelty of this approach lies in the combination of this information extracted from the document image in real time just prior to real world exploration with dynamic visual perception. An artificial neural system integrates both sets of information by having the robot recognize navigation signs from the floor plan and then look for them in the expected sequence in the environment. Thereafter, while determining and executing appropriate movements as a response to the environment, the robot links and learns to associate the undertaken movements to the respective signs. The analysis of the floor plan involves a detailed process permitting the robot to extract the required information in real time. This process is elaborated in this paper, along with experiments with the robot using the proposed architecture in a simple navigation scenario, hence proving the feasibility of the approach.

With the maturity of machine learning methods to provide satisfying Content-Based Image Retrieval systems (CBIR), research focus has recently turned back towards visual saliency analysis. The goal in these works is to extract even more efficient visual features than the existing ones. However, analyzing visual saliency is critically dependent on the task to be accomplished from the extracted visual features. A significant number of CBIR systems consider image retrieval as a binary classification problem: what is relevant for the user against what is irrelevant. In this paper, we focus on extracting relevant gaze features within the paradigm of visual preference in order to support the annotation by gaze for a CBIR system. We thus define a gaze acquisition protocol, design a benchmark from a subset of Pascal VOC database and present an in depth analysis of eye-tracking data for visual preference paradigm. Our paper provides new informations on relevant gaze features for image binary classification.

International audience

Content based image retrieval (CBIR) has been the center of interest for a long time. A lot of research have been done to enhance the performance of these systems. Most of the proposed works focused on improving the image representation(bag-of-features) and classification methods. In this paper, we focus on enhancing the second component of CBIR system: region appearance description method. In this context, we propose a new descriptor describing the spatial frequency property of some perceptual features in the image. This descriptor has the advantage of being lower dimension vs. traditional descriptors as SIFT (60 vs. 128), thus computationally more efficient, with only 5% loss in performance using a typical CBIR algorithm on VOC 2007 dataset.

We present eBDtheque, a database of various comic book images and their ground truth for panels, balloons and text lines plus semantic annotations. The database consists of a hundred pages of various comic book albums, Franco-Belgian, American comics and mangas. Additionally, we present the piece of software used to establish the ground truth and a tool to validate results against this ground truth. Everything is publicly available for scientific use on http://ebdtheque.univ-lr.fr.

In this paper, we present a long term learning system for content based image retrieval over a network. Relevant feedback is used among different sessions to learn both the similarity function and the best routing for the searched category. Our system is based on mobile agents crawling the network in search of relevant images. An ant-behavior algorithm is used to learn the category dependent routing. With experiments on trecvid'05 key-frame dataset, we show that the smart association of category dependent routing and active learning leads to an improvement of the quality of the retrieval over time.

In this article, we present a framework for distributed content based image retrieval with online learning based on ant-like mobile agents. Mobile agents crawl the network to find images matching a given example query. The images retrieved are shown to the user who labels them, following the classical relevant feedback scheme. The labels are used both to improve the similarity measure used for the retrieval and to learn paths leading to sites containing relevant images. The relevant paths are learned in an ethologically inspired way. We made experiments on the trecvid 2005 keyframe dataset showing that learning both the similarity function and the localization of the relevant images leads to a significant improvement. We also present an extension with the reuse of learned paths for later sessions leading to a further improvement.

Information retrieval techniques have to face both the growing amount of data to be processed and the "natural"' distribution of these data over the network. Hence, we introduce in this paper a new architecture for image retrieval in distributed image databases, based on multi-agent, systems. Our system, inspired by "ant-agents", uses labels provided by the user for learning both the searched category of images and die path to the most relevant databases. We then show how effective can be our architecture on a generalist image database network.

In this article, Web-search mobile agents inspired by ethology and robotics are presented. 2 strategies are compared: 1/sup st/ uses pheromones to mark the route to a given goal; 2/sup nd/ involves a cognitive map to internalize a representation of the Web. The implementation on a set of FI-PAOS platforms is given and unintuitive results concerning agent mobility are given.

In this letter we describe a hippocampo-cortical model of spatial processing and navigation based on a cascade of increasingly complex associative processes that are also relevant for other hippocampal functions such as episodic memory. Associative learning of different types and the related pattern encoding-recognition take place at three successive levels: (1) an object location level, which computes the landmarks from merged multimodal sensory inputs in the parahippocampal cortices; (2) a subject location level, which computes place fields by combination of local views and movement-related information in the entorhinal cortex; and (3) a spatiotemporal level, which computes place transitions from contiguous place fields in the CA3-CA1 region, which form building blocks for learning temporospatial sequences. At the cell population level, superficial entorhinal place cells encode spatial, context-independent maps as landscapes of activity; populations of transition cells in the CA3-CA1 region encode context-dependent maps as sequences of transitions, which form graphs in prefrontal-parietal cortices. The model was tested on a robot moving in a real environment; these tests produced results that could help to interpret biological data. Two different goal-oriented navigation strategies were displayed depending on the type of map used by the system. Thanks to its multilevel, multimodal integration and behavioral implementation, the model suggests functional interpretations for largely unaccounted structural differences between hippocampo-cortical systems. Further, spatiotemporal information, a common denominator shared by several brain structures, could serve as a cognitive processing frame and a functional link, for example, during spatial navigation and episodic memory, as suggested by the applications of the model to other domains, temporal sequence learning and imitation in particular.

A neural network model of intrahippocampal and hippocampo-cortico-gangliobasal loops allows the robotic implementation of the spatial information contained within cognitive maps in neural space, into temporo-spatial sequences of movements during goal-oriented navigation in outer space. At the representation level, the intrahippocampal loop features place cells in entorhinal cortex and transition cells in CA3-CA1 which constitute, as a population, spatial and contextual maps, respectively. Path integration converges with place cell information in the subiculum. The spatial representation in deep and superficial layers of the entorhinal cortex are dissociated; the unidirectional connections between these two layers close the intrahippocamal loop. The prefrontal cortex, at the junction between representation and implementation, receives from three hippocampal subsystems and stores a global graph-map of an environment and the goal locations on this map. The diffusion of activation from active goals through the graph allows path selection in the motivational limbic prefrontal cortex and planning in the executive lateral part. The top-down output from prefrontal cortex and the bottom-up output from the hippocampus combine onto the accumbens the first stage for the stepwise selection and implementation of the optimal actions in the direction of the goal. Proactive and reactive functioning modes are dissociated.

We present a Web-search ant-agent inspired by ethology and robotics. We detail its implementation on a set of FIPAOS platforms and show useful results in route finding and rerouting. Finally, we discuss its interest and drawbacks in comparison with classical search engines and give perspectives to overcome them.

Visually guided landmark navigation is based on space coding by hippocampal place cells (PC). A biologically realistic architecture of cooperative-competitive associative networks (implemented as a control system for mobile agents) emulates PC activity during local navigation in exploration and goal-retrieval. The system builds and stores panoramic views from landmarks and compares them with current inputs. Mismatch-induced low levels of recognition trigger a vigilance burst, which favors either the recognition of an alternative place category or the creation of a new category. The sole implementation of visual "What" and "Where" information does not restrain the generality of the model since several modalities could cooperate to give rise to more robust place field spatial categories. Providing the system with real visual inputs automatically extracted from a natural environment demonstrates that interspecies differences in PC coding result more from characteristics of the visual systems than from differences in processing. Conversely, differences in PC multiple codes within a system result from different levels of processing and/or different degrees of multimodality. Each code could be used within different navigational strategies. A control system derived from the model allows a mobile agent to learn a few places and the associated actions required to reach a goal. Generalisation property of the model provides the capacity to join the goal from any place in the learned environment.

We describe how a mobile robot controlled only by visual information can retrieve a particular goal location in an open environment. Our model does not need a precise map nor to learn all the possible positions in the environment. The system is a neural architecture inspired from neurobiological studies using the recognition of visual patterns called landmarks. The robot merges this visual information and its azimuth to build a plastic representation of its location. This representation is used to learn the best movement to reach the goal. A simple and fast online learning of a few places located near the goal allows the robot to reach the goal from anywhere in its neighborhood. The system uses only an egocentric representation of the robot environment and presents very high generalization capabilities. We describe an efficient implementation tested on our robot in two real indoor environments. We show the limitations of the model and its possible extensions to create autonomous robots only guided by visual information.

In this paper we describe how a mobile robot can autonomously learn and "recognize" simple objects present somewhere in an indoor visual scene. The experiment involves transposing a classical conditioning experiment on a mobile robot. We propose the use of a selective attention mechanism to reduce the amount of computation involved by the complete image analysis. Objects are categorized according to their associated actions that are learned in accordance with a reward/punishment procedure. Our approach emphasizes the importance of a movement reflex mechanism based on the use of the same egocentric representation from the visual information to the motor output. Finally, we highlight the impact of information coding in self organised topological maps on the robot performances.