Visual navigation

In this paper, we present a multi-sensor fusion based monocular
visual navigation system for a quadrotor with limited payload,
power and computational resources. Our system is equipped with an
inertial measurement unit (IMU), a sonar and a monocular
down-looking camera. It is able to work well in GPS-denied and
markerless environments. Different from most of the keyframe-based
visual navigation systems, our system uses the information from
both keyframes and keypoints in each frame. The GPU-based speeded
up robust feature (SURF) is employed for feature detection and
feature matching. Based on the flight characteristics of
quadrotor, we propose a refined preliminary motion estimation
algorithm combining IMU data. A multi-level judgment rule is then
presented which is beneficial to hovering conditions and reduces
the error accumulation effectively. By using the sonar sensor, the
metric scale estimation problem has been solved. We also present
the novel IMU+3P (IMU with three point correspondences) algorithm
for accurate pose estimation. This algorithm transforms the 6-DOF
pose estimation problem into a 4-DOF problem and can obtain more
accurate results with less computation time. We perform the
experiments of monocular visual navigation system in real indoor
and outdoor environments. The results demonstrate that the
monocular visual navigation system performing in real-time has
robust and accurate navigation results of the quadrotor.

Citation: Wei Zheng, Fan Zhou, Zengfu Wang. Robust and accurate monocular visual navigation combining IMU for a quadrotor. IEEE/CAA Journal of Automatica Sinica, 2015,  2(1): 33-44

Geographic information (also called geo-information, spatial information or geospatial information) plays an increasingly important role in our society. Location Based Services, applications for urban planning, disaster management systems rely on geo-data and in many cases up-to-date geo-information is crucial, e.g. in disaster response systems. In this paper we focus on a special kind of geo-information: 3D spatial data to be used in 3D GIS (Geographic Information System), CAD (Computer Aided Design), Virtual Reality or Augmented Reality applications. Because the production of up-to-date and accurate geo-information is very costly, exchanging and sharing data between organizations is often the only way to get all relevant information to decision makers and other end users. In principle Internet has created new possibilities for the shared use of geo-data. Data sets do not have to be copied anymore on CD's or tape, but could be accessed online, via Web services. Progress towards shared use of geo-data via the Web is slow however. One of the bottlenecks is the heterogeneous nature of geographic data, both in a technical sense (data format) and in a semantic sense (data model and meaning). Interoperability (of data and software) is therefore a key issue that needs to be resolved. In this paper we will look at the interoperability initiatives of the OpenGIS Consortium (OGC). One of the main principles of OGC is, that in order to improve data and software interoperability, it is necessary to create standardized interfaces between the different components (possibly of different vendors) of geo-information systems. We will investigate possibilities to publish 3D geo-data on the Web using OpenGIS compliant Web services. We will report on a 3D Web Mapping prototype environment built at our research institute. One purpose of the prototype is to see whether the present OGC specifications are adequate for access to 3D geo-data (query, visualization, navigation and analysis). Apart from the OpenGIS Web service specifications also another emerging Web standard is used in the prototype: X3D (eXtensible 3D) of the Web3D Consortium.

The Visual Semantic Web (ViSWeb) paradigm enhances human accessibility to the current Semantic Web technology by enabling the visualization of knowledge. Arguing against the claim that humans and machines need to look at different knowledge representation formats, Object-Process Methodology (OPM) is shown to enable modeling of systems in a single graphic and textual model. ViSWeb provides for representation of knowledge over the Web in a unified way that caters to humans as well as machines. ViSWeb is developed as an OPM-based layer on top of XML/RDF/OWL to express knowledge visually and in natural language. Both the graphic and the textual representations are strictly equivalent. Being intuitive yet formal, they are not only understandable to humans, but are also amenable to computer processing. The advantages of the ViSWeb approach include equivalent graphic-text knowledge representation, visual navigability, semantic sentence interpretation, specification of system dynamics, and complexity management. The ability to use such bimodal knowledge representation that is both human understandable and machine processable is a major step forward in the evolution of the Semantic Web.

Abstract— In this paper, we present a controller for visual navigation, which utilizes a time-independent varying reference in the feedback,law. The navigation,framework,relies on a monocular camera, and the path is represented as a series of key images. The varying reference is determined,using a vector field, derived from the previous and next key images. Results in a simulated environment, as well

In untraditional environments and unexpected situations, it is not always possible for a fixed-architecture robot to meet all the task requirements. Self-assembly enables free bodies to autonomously join together and complete a task that would otherwise not be achievable with the independent modules. The main theme of this thesis is to investigate the possibility of converting frail and inflexible mobile robots to a single flexible modular robot, and conversely, to split up a long and slow structure into separate fast modules. To prove the possibility of having such a system, we propose a novel connection mechanism, as well as a vision-based docking guidance system. This docking system along with a parking control algorithm allow single articulated- steering mobile robots to autonomously connect and disconnect. So, the concept of this work lies at the intersection between modular and wheeled mobile robotic systems. The goal is to improve the performance of mobile robots by autonomously changing their locomotion method to best fit the terrain conditions

Visual processing is very important for robot navigation. It has been demonstrated that many complex operations, which deserve an intelligent behaviour, can be performed relying only on re exes to visual stimuli. In this framework the detection of corridors of free space along the robot trajectory is certainly a very important capability to safely navigate. Stereo vision and motion parallax can be used as cues to infer scene structure and determine free space areas. In this paper we propose a cooperative schema in which binocular disparity, computed on several stereo images over time, is combined with optical ow from the same sequence to obtain a relativedepth map of the scene. Both time-to-impact and depth scaled by the distance of the camera from the xation point in space are considered as good, relative measurements which are based on the viewer (but centered on the environment). Simple relations are de ned which combine disparity and optical ow for relative-depth estimates. In particular, some aspects of dynamic stereo vision are considered. Two experiments, performed on image sequences from real scenes, are presented.

Figure 1: Fixation maps over an art work that contains separate panels, which, when viewed in the intended order, tell a coherent story. In this work a gaze direction technique is used to visually guide the viewer through the image. As can be seen above the distribution of fixations is very different for static viewing (A), and Gaze Directed viewing (B). The highlighted rectangular regions indicate story panels, and the numbered circles indicate viewing order. In the gaze directed case, both order viewing accuracy and fixations on relevant panels increased.

Navigation using visual aids has always been an important skill for any navigator. One of the most important visual aids on land and along the coast is landmark. Traditional cartography has developed many techniques in representing landmarks on a map, including hill shaind, symbology, and perspective drawings. One of the problems in using a traditional map, which is an orthogonal projection of the land features, is that its "air-perspective" does not match the "groundperspective" of a navigator on land. In this paper, we will discuss the role of landmarks in navigation, the use of photographic images to represent landmarks, and the development of a prototype for visual navigation using a multi-media map.

Spatial databases contain feature classes that represent real world. Content of geographic database is mainly designed for GIS analysis. However, a spatial database for cartographic visualization should contain some more rules and information for visualization. Cartographic visualization of a spatial database is usually in map format. Being expressed in this study, Joint Operations Graphics (JOG) is cartographic visualization of Vector

where he is currently an assistant professor. In 1986, he was a research assistant at the Department of Computer Science of Trinity College, Dublin, Ireland, where he was developing a system for the analysis of image data primarily aimed at the investigation of low-level visual proceses. In 1989, he was a visiting scientist at Thinking Machines Co. developing parallel algorithms for dynamic image processing on the Connection Machine. Dr. Tistarelli's research interests include robotics, computer vision (particularly in the area of three-dimensional and dynamic scene analysis), image processing, and artificial intelligence.

This diploma thesis discusses the impact of authoring systems on multidimensional documents. A first part illustrates concepts on multidimensional documents and how to represent them. Another issue is navigation through n-dimensional documents. The second part of the work implements a prototype of a multidimensional authoring system, named MDAT (Multidimensional Authoring Tool). MDAT helps teachers in creating their own interactive courses on the Web. Students, at the other hand, are supported with a navigational aide to keep orientation in these documents.

Narrative art tells a story, either as a moment in an ongoing story or as a sequence of events unfolding over time. In many works of art separate panels within the same frame are used to depict the sequence of events. Often, there is no clear delineation between these panels, or any indication of the optimal viewing order. To improve

Mobile robot localization from large-scale appearance mosaics has been showing increasing promise as a low-cost, high-performance and infrastructure-free solution to vehicle guidance in man-made environments. The feasibility of this technique relies on the construction of a locally smooth and globally consistent high-resolution mosaic of the vehicle's environment, efficiently done using observations that have low spatial and temporal persistence. The problem of loop closure in cyclic environments that plagues this process is one that is commonly encountered in all map-building procedures, and its solution is often computationally expensive. This paper presents a method that reliably generates consistent maps at low computational cost, while fully exploiting the topology of the observations. Extensions to a realtime implementation are discussed along with results using simulated data and those from real indoor environments.

This paper describes a system for structure-from-motion using vanishing points and three-dimensional lines extracted from omni-directional video sequences. To track lines, we use a novel dynamic programming approach to improve ambiguity resolution, and we use delayed states to aid in the initialization of landmark. By reobserving vanishing points we get direct measurements of the robot's 3-D attitude that are independent of its position. Using vanishing points simplifies the representation since parallel lines share the same direction states. We show the performance of the system in various indoor and outdoor environments and include comparisons with independent 2-D reference maps for each experiment.

Figure 1: Fixation maps over an art work that contains separate panels, which, when viewed in the intended order, tell a coherent story. In this work a gaze direction technique is used to visually guide the viewer through the image. As can be seen above the distribution of fixations is very different for static viewing (A), and Gaze Directed viewing (B). The highlighted rectangular regions indicate story panels, and the numbered circles indicate viewing order. In the gaze directed case, both order viewing accuracy and fixations on relevant panels increased.

In order to be fully autonomous, a mobile robot must possess the skill of finding its location in a particular environment. In other words, while navigating, a mobile robot must be capable of finding its location in a map of the environment (i.e. its pose < x, y, θ >), otherwise the robot will not be able to complete its task. The problem becomes specially challenging if the robot does not possess any external measure of its global position. Naively, the position/orientation of the robot can be determined using odometry sensors or inertial systems. However, these sensors lack of accuracy when used for long periods of time, due to wheel slippage, drifts and other problems. Localization techniques are used instead, in order to find the position of the mobile agent in the space. The great majority of localization methods rely on finding salient characteristics sensed by the robot and relating them with a map of the environment. These salient characteristics of space are usually c...

We introduce a method for autonomous robot navigation based on homographies computed between current image and images taken in a previous teaching phase with a monocular vision system. The features used to estimate the homography are vertical lines automatically extracted and matched. From homography, the underlying motion correction between the reference path and the current robot location is computed. The proposed method, which uses a sole calibration parameter, has turned out to be specially useful to correct heading and lateral displacement, which are critical in systems based on odometry. We have tested the proposal in simulation, and with real images. Besides, the visual system has been integrated into an autonomous wheelchair for handicapped, working in real time with robustness.

The electrochemical characteristics of L605 alloy, with various surface finishes obtained after electrolytic polishing, were investigated. This Co-Cr alloy has a clinical history in dental and orthopaedic implants and more recently in cardiovascular stent applications, where its high elastic modulus and radiopacity are desirable properties. Radiopaque materials are advantageous as they allow for better visual navigation in minimal invasive implantation procedures. For applications in the human body, surface corrosion resistance is the most important characteristic to evaluate the biocompatibility. Several studies of the corrosion resistance of Co-Cr alloy biomaterials in simulated human body fluids have been reported. The paper presents recent achievements in increasing corrosion resistance of L605 alloy after the process of electrolytic polishing in the presence of a magnetic field (MEP).

The Visual Semantic Web (ViSWeb) is a new paradigm for enhancing the current Semantic Web technology. Based on Object-Process Methodology (OPM), which enables modeling of systems in a single graphic and textual model, ViSWeb provides for representation of knowledge over the Web in a unified way that caters to human perceptions while also being machine processable. The advantages of the ViSWeb approach include equivalent graphic-text knowledge representation, visual navigability, semantic sentence interpretation, specification of system dynamics, and complexity management. Arguing against the claim that humans and machines need to look at different knowledge representation formats, the principles and basics of various graphic and textual knowledge representations are presented and examined as candidates for ViSWeb foundation. Since OPM is shown to be most adequate for the task, ViSWeb is developed as an OPMbased layer on top of XML/RDF/OWL to express knowledge visually and in natural language. Both the graphic and the textual representations are strictly equivalent. Being intuitive yet formal, they are not only understandable to humans but are also amenable to computer processing. The ability to use such bimodal knowledge representation is potentially a major step forward in the evolution of the Semantic Web.

In this paper we describe how a mobile robot controled 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 those visual information and their 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 on line 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 egocentric representation of the robot environment and presents very high generalization capabilities. We describe an e cient implementation tested on our robot in two real indoor environments. We show the limitations of the model and its possible extensions to create autononous robots only guided by visual information.

In this paper, we highlight the main research directions currently pursued by the investigators for the development of new tools to improve Web accessibility for users with visual disabilities. The overall principle is to create intelligent software agents used to assist visually impaired individuals in accessing complex on-line data organizations (e.g., tables, frame structures) in a meaningful way. Accessibility agents make use of knowledge representation structures (automatically or manually derived) to assist users in developing navigation plans; these are employed to locate given pieces of information or to answer user's desired goals.

A visual navigation system is described which uses texture mapped primitives to represent clusters of objects to maintain high and approximately constant frame rates. In cases where there are more unoccluded primitives inside the viewing frustum than can be drawn in real-time on the workstation, this system ensures that each visible object, or a cluster that includes it, is drawn in each frame. The system supports the we of traditional "level-of-detail" representations for individual objects, and supports the automatic generation of a certain type of level-of-detail for objects and clusters of objects. The concept of choosing a representation from among those associated with an object that accounts for the direction from which the object is viewed is also supported. The level-of-detail concept is extended to the whole model and the entire scene is stored as a hierarchy of levels-of-detail that is traversed top-down to iind a good representation for a given viewpoint. This system does not assume that visibility information can be extracted from the model and is thus especially suited for outdoor environments.

Spatial databases contain feature classes that represent real world. Content of geographic database is mainly designed for GIS analysis. However, a spatial database for cartographic visualization should contain some more rules and information for visualization. Cartographic visualization of a spatial database is usually in map format. Being expressed in this study, Joint Operations Graphics (JOG) is cartographic visualization of Vector Smart Map Level 1 (VMAP L-1) spatial database. JOG Air and Ground series at scale 1:250000 are used for joint operations of allied forces. VMAP L-1 is a spatial database with a content of middle scale maps such as JOGs at scale 1:250000. JOG Air and Ground series are used for planning of short and long distance flight, air-ground tactical operations, helicopter operations, tactical and close-air support, visual navigation, and air-ground coordinated operations and strategic/operational planning purposes by ground units and civil aviation. First and se...

Narrative art tells a story, either as a moment in an ongoing story or as a sequence of events unfolding over time. In many works of art separate panels within the same frame are used to depict the sequence of events. Often, there is no clear delineation between these panels, or any indication of the optimal viewing order. To improve

Resumo-Este trabalho apresenta resultados parciais do Projeto "Sistemas Virtuais de Ensino e Treinamento Baseados na Internet para Suporte à área da Saúde" (MCT/FINEP-Ação Transversal-Software, Edital 06/2004). Nesta perspectiva, busca-se (i) prover um ambiente de estruturas anatômicas através de visualização, navegação e interação em terceira dimensão e (ii) criar um sistema tutor para educação médica para aprendizagem de anatomia humana. Neste primeiro momento, enfatizamos o domínio do crânio humano. Este trabalho insere-se na área de educação médica via Sistemas Tutores Inteligentes com ontologias como formalismo de representação do domínio da aplicação.

This paper describes a generic approach to the design of a 3D virtual environment for visual navigation. The basic design principle is that users need to understand how an environment is organised and how they can utilise the knowledge to find their way through the environment. In this paper, we focus on the relationship between implicit semantic structures associated with a collection of domain-specific documents and users' cognitive needs in the context of visual navigation. We extend the notion of Pathfinder networks in order to represent implicit semantic structures in a 3D virtual environment. Users' cognitive needs in visual navigation are conceptualised based on the concept of cognitive maps.

In this paper, we present an approach for the design and control of both reflexive and purposive visual tasks. The approach is based on the bidimensional appearance of the objects in the environment and explicitly takes into account independent object motions. A linear model of camera-object interaction is embedded in the control scheme, which dramatically simplifies visual analysis and control by reducing the size of visual representation. We describe the implementation of three visual tasks of increasing complexity, obtained with the proposed scheme and based on active contour analysis and polynomial planning of image contour transformations. Both simulations and real-time experiments with a robotic eye-in-hand configuration are discussed, validating the approach in terms of robustness and applicability to visual navigation, active exploration and perception, and human-robot interaction.

We address the problem of comprehending cause and effect relationships between relatively independent behavior components of a single application. Our focus is on the paradigm of behavioral, scenario-based, programming, as captured by the language of live sequence charts (LSC) or its Java-based counterpart, BPJ. In this programming paradigm, multi-modal behaviors can be specified separately, and are integrated only at run time. We present a tool, with which the user can easily follow the decisions of the collective execution mechanism. It shows the behaviors and events that were executed at each point in time, and those that were delayed or abandoned, as well as the causes and reasons behind these run-time choices. The dynamic effects of such decisions on the system's behavior can be seen easily too.

The use of a contrasting line for the visual navigation of autonomous mobile robots in a factory environment is developed. Minimum and maximum linewidths are determined analytically by considering sensor geometry, field of view, and error conditions present in the system. The effects of these error conditions on the width of the line, as seen in the image plane, determines the optimal linewidth. Numerical examples using typical sensor parameters are given.

Computer vision is much more than a technique to sense and recover environmental information from an UAV. It should play a main role regarding UAVs' functionality because of the big amount of information that can be extracted, its possible uses and applications, and its natural connection to human driven tasks, taking into account that vision is our main interface to world understanding. Our current research's focus lays on the development of techniques that allow UAVs to maneuver in spaces using visual information as their main input source. This task involves the creation of techniques that allow an UAV to maneuver towards features of interest whenever a GPS signal is not reliable or sufficient, e.g. when signal dropouts occur (which usually happens in urban areas, when flying through terrestrial urban canyons or when operating on remote planetary bodies), or when tracking or inspecting visual targets -including moving ones-without knowing their exact UMT coordinates. This paper also investigates visual servoing control techniques that use velocity and position of suitable image features to compute the references for flight control. This paper aims to give a global view of the main aspects related to the research field of computer vision for UAVs, clustered in four main active research lines: visual servoing and control, stereo-based visual navigation, image processing algorithms for detection and tracking, and visual SLAM. Finally, the results of applying these techniques in several applications are presented and discussed: this study will encompass power line inspection, mobile target tracking, stereo distance estimation, mapping and positioning.

This chapter presents an introduction to current research devoted to the visual servoing problem of guiding differentially driven robots, more specifically, unicycle-like vehicles, taking into consideration limited field-of-view constraints. The goal is to carry out accurate servoing of the vehicle to a desired posture using only feedback from an on-board camera. First, a position based scheme is proposed, adopting a hybrid control law to cope with limited camera aperture. This scheme relies on a localization method based on extended Kalman filter technique that takes into account the robot motion model and odometric data. To increase the potentiality of the visual servoing scheme with respect to existing solutions, which achieve similar goals locally (i.e., when the desired and actual camera views are sufficiently similar), the proposed method visually navigate the robot through an extended visual map before eventually reaching the desired goal. The map construction is part of the approach proposed here, which is then called Visual SLAM for Servoing. Position based scheme accuracy are intrinsically related to the effectiveness of the localization process, which is related to the estimation of 3D information on both the robot and the environment. A shortcut overcoming the estimation process uses visual information directly in the image domain. In this spirit, an image based scheme is presented. The controller is devoted to constantly track desired image feature trajectories. Such trajectories represent optimal (shortest) paths for the vehicle from the 3D initial position towards the desired one. Optimal trajectories satisfies

Abstract. In this paper, we propose an autonomous precision landing method for an unmanned helicopter based on an on-board visual navigation system consisting of a single pan-tilting camera, off-the-shelf computer hardware and inertial sensors. Compared to existing methods, the system doesn’t depend on additional sensors (in particular not on GPS), offers a wide envelope of starting points for the autonomous approach, and is robust to different weather conditions. Helicopter position and attitude is estimated from images of a specially designed landing pad. We provide results from both simulations and flight tests, showing the performance of the vision system and the overall quality of the landing. 1

The paper describes a fuzzy perception based guidance of an autonomous ground vehicle. The technique proposed uses a visual servoing approach in which the control incorporates directly the visual feedback. The application involves guiding the vehicle through roadways by detecting the road edges from the image space. The vehicle and the scene models are established from the system behavior and the visually perceived features. The method is based on the fuzzy perception of the road edges from the gradient orientation image. In our work, the problem of road edge detection is viewed as a phenomena of perceiving gradient direction levels and then tracing the locus of the vectors which correspond to dominant linear features. In order to reinforce the detection process, the edge perception operator is formulated as a membership function applied on the gradient orientation image in the intensity domain. With this fusion process, the conventional number of edge levels are found to be reduced...

This paper presents an algorithm that uses visual information to achieve the homing of an autonomous agent inside a previously visited environment. An image grabbed at the target position is compared with the currently perceived one to determine the relative position of the robot and of its target. Only particular regions of the image, called Visual References, are taken into account. A Visual References correlation criterion that uses the Fourier-Mellin Transform to match the Visual References in different images is employed. This transform in fact allows computing Visual References that are invariant to rotation, scaling and translation. Robustness due to the use of the Mellin Transform in the Visual References selection and coupling leads to more precise navigation. Tests and results are presented. © 2002 Elsevier Science B.V. All rights reserved.

Visual processing is very important for robot naviga tion. In this framework the detection of corridors of free space along the robot trajectory is certainly a very important capability to safely navigate. Stereo vision and motion parallax can be used as cues to infer scene structure and determine free space areas. In this paper we propose a cooperative schema in which binocular disparity, computed on several stereo images over time, is combined with optical flow from the same sequence to obtain a relative-depth map of the scene. Both time-bimpact and depth scaled by the distance of the camera from the fixation point in space are considered as good, relative measurements which are based on the viewer (but centered on the environment). Two experiments, performed on image sequences from real scenes, are presented.

For autonomous vehicles to achieve terrain navigation, obstacles must be discriminated from terrain before any path planning and obstacle avoidance activity is undertaken. In this paper, a novel approach to obstacle detection has been developed. The method finds obstacles in the 2D image space, as opposed to 3D reconstructed space, using optical flow. Our method assumes that both nonobstacle terrain

In this paper we address the problem of visual navigation of a mobile robot which simultaneously obtains metric localization and scene reconstruction using homographies. Initially, the robot is guided by a human and some scenes during the trip are stored from known reference locations. The interest of this paper consist in the possibility of getting real and precise data of the robot motion and the scene, which presents some advantages over other existing approaches. For example, it allows the robot to carry out other trajectories than the executed during the teaching phase. We show an extensive analysis of the output in presence of errors in some of the inputs.

Comprehensive data analysis has become indispensable in a variety of environments. Standard On-Line Analytical Processing (OLAP) systems, designed for satisfying the reporting needs of the business, tend to perform poorly or even fail when applied in non-business domains such as medicine, science, or government. The underlying multidimensional data model is restricted to aggregating only over summarizable data, i.e. where each dimensional hierarchy is a balanced tree. This limitation, obviously too rigid for a number of applications, has to be overcome in order to provide adequate OLAP support for novel domains. We present a framework for querying complex multidimensional data, with the major effort at the conceptual level as to transform irregular hierarchies to make them navigable in a uniform manner. We provide a classification of various behaviors in dimensional hierarchies, followed by our two-phase modeling algorithm that proceeds by eliminating irregularities in the data with subsequent transformation of complex hierarchical schema into a set of well-behaved sub-dimensions. Mapping of the data to a visual OLAP browser relies solely on metadata which captures the properties of facts and dimensions as well as the relationships across dimensional levels. Visual navigation is schemabased, i.e. users interact with dimensional levels and the data instances are displayed on-demand. The power of our approach is exemplified using a real-world study from the domain of academic administration.

This paper presents an measure for image similarity based on local feature descriptions and geometric constraints. We show that on the basis of this similarity an appearance graph representation of the environment of a mobile robot can be made. This graph can be used for representing semantic information about the space, and can be used for visual navigation. The image similarity measure is robust for occlusions by people in the neighbourhood of the robot.

Resumo -Este trabalho apresenta resultados parciais do Projeto "Sistemas Virtuais de Ensino e Treinamento Baseados na Internet para Suporte à área da Saúde" (MCT/FINEP -Ação Transversal -Software, Edital 06/2004). Nesta perspectiva, busca-se (i) prover um ambiente de estruturas anatômicas através de visualização, navegação e interação em terceira dimensão e (ii) criar um sistema tutor para educação médica para aprendizagem de anatomia humana. Neste primeiro momento, enfatizamos o domínio do crânio humano. Este trabalho insere-se na área de educação médica via Sistemas Tutores Inteligentes com ontologias como formalismo de representação do domínio da aplicação.

Graphic and metric archaeological documentation is an activity that requires the capture of information from different sources, accurate processing and comprehensive analysis. If monitoring of the state of conservation is required, this task has to be performed before intervention, during and after the completion of the works in a repetitive way.

This article presents a technique for modeling and calibrating a camera with integrated low-cost iner- tial sensors, three gyros and three accelerometers for full 3D sensing. Inertial sensors attached to a camera can provide valuable data about camera pose and movement. In biological vision systems, inertial cues provided by the vestibular system, are fused with vision at an early processing

This article presents a technique for modeling and calibrating a camera with integrated low-cost inertial sensors, three gyros and three accelerometers for full 3D sensing. Inertial sensors attached to a camera can provide valuable data about camera pose and movement. In biological vision systems, inertial cues provided by the vestibular system, are fused with vision at an early processing stage. Vision systems in autonomous vehicles can also benefit by taking inertial cues into account. Camera calibration has been extensively studied, and standard techniques established. Inertial navigation systems, relying on high-end sensors, also have established techniques. Nevertheless, in order to use off-the-shelf inertial sensors attached to a camera, appropriate modeling and calibration techniques are required. For inertial sensor alignment, a pendulum instrumented with an encoded shaft is used to estimate the bias and scale factor of inertial measurements. For camera calibration, a standard and reliable camera calibration technique is used, based on images of a planar grid. Having both the camera and the inertial sensors calibrated and observing the vertical direction at different poses, the rigid rotation between the two frames of reference is estimated, using a mathematical model based on unit quaternions. The technique for this alignment and consequent results with simulated and real data are presented at the end of this article.