The future of autonomous driving is slowly approaching, but there are still many steps to take before it can become a reality. It is crucial to pay attention to road infrastructure, because without it, intelligent vehicles will not be able to operate reliably, and it will never be possible to dispense of driver’s control. This paper presents the work carried out for the detection of road markings damage using computer vision techniques. This is a complex task for which there are currently not many papers and large image sets in the literature. This study uses images from the public Road Damage Detection dataset for the D44 defect and also provides 971 new labelled images for Spanish roads. For this purpose, three detectors based on deep learning architectures (Faster RCNN, SDD and EfficientDet) have been used and single-source and mixed-source models have been studied to find the model that best fits the target images. Finally, F1-score values reaching 0.929 and 0.934 have been obtained for Japanese and Spanish images respectively which improve the state-of-the-art results by 25%. It can be concluded that the results of this study are promising, although the collection of many more images will be necessary for the scientific community to continue advancing in the future in this field of research.

This paper proposes a novel approach to registering deformations of 3D non-rigid objects for Augmented Reality applications. Our prototype is able to handle different types of objects in real-time regardless of their geometry and appearance (with and without texture) with the support of an RGB-D camera. During an automatic offline stage, the model is processed in order to extract the data that serves as input for a physics-based simulation. Using its output, the deformations of the model are estimated by considering the simulated behaviour as a constraint. Furthermore, our framework incorporates a tracking method based on templates in order to detect the object in the scene and continuously update the camera pose without any user intervention. Therefore, it is a complete solution that extends from tracking to deformation formulation for either textured or untextured objects regardless of their geometrical shape. Our proposal focuses on providing a correct visual with a low computational cost. Experiments with real and synthetic data demonstrate the visual accuracy and the performance of our approach.

Deformable models are widely used in many disciplines such as engineering and medicine. Real objects are usually scanned to create models in such applications. In many cases the shape of the object is extracted from volumetric data acquired during the scanning phase. At the same time, this volume can be used to define the model's appearance. In order to achieve a visualization that unifies the shape (physical model) and appearance (scanned volume) specially adapted volume rendering techniques are required. One of the most common volumetric visualization techniques is ray casting, which also enables the use of different corrections or improvements such as adaptive sampling or stochastic jittering. This paper presents an extensive study about a ray casting method for tetrahedral meshes with an underlying structured volume. This allows a direct visualization of the deformed model without losing the information contained in the volume. The aim of this study is to analyse and compare the different methods for ray traversal and illumination correction, resulting in a comprehensive relation of the different methods, their computational cost and visual performance.

A real-time Augmented Reality system that assists in maintenance is presented (Figure 1). The novelty of the proposed system relies on its ability of being a complete framework that is self supplied. All the data is extracted automatically during an offlne phase from a collection of 3D parts that are provided as untextured 3D triangle meshes. Our disassembly planning module computes the assembly/disassembly sequence finding collision-free trajectories [1]. Moreover, some basic geometric features, such as edges and junctions, are extracted to address the tracking problem. Additionally, these geometric features let us perform the 3D model recognition, which solves the first camera pose problem and performs the recovery in case of tracking failure [3]. During the online phase, these geometric features are used to build a markerless tracking that updates continuously the camera pose from a monocular image. This markerless tracking combines an edge tracker, a point based tracker and a 3D particle filter to offer a robust 3D tracking against undesirable conditions [2]. Once the camera pose has been computed, the system offers an augmented instruction, indicating how to assemble/disassemble the next part. Thus, the proposed AR system is a complete framework that runs in real-time and tries to facilitate the work of workers, replacing the paper-based documentation.

ABSTRACT Marker systems are a widely used optical tracking method that does not support occlusions. Thus, this paper proposes a new marker design to overcome the problem of marker occlusions. It is highly adaptable, because it can be used by any marker tracking system that uses its central area to codify the digital identification. Our proposal takes advantage of an untapped frame to place some textures that will be tracked during marker occlusion. In addition, these textures are customizable, which lets users make their own designs. Two tracking methods are combined to offer a robust tracking, updating the six degrees of freedom of the camera in real time. The first one is a fast technique based on temporal coherence, whereas the second one is a robust technique based on appearance, which is used as a recovery mode. Copyright © 2012 John Wiley & Sons, Ltd.

This paper proposes a new real-time Augmented Reality based tool to help in disassembly for maintenance operations. This tool provides workers with augmented instructions to perform maintenance tasks more efficiently. Our prototype is a complete framework characterized by its capability to automatically generate all the necessary data from input based on untextured 3D triangle meshes, without requiring additional user intervention. An automatic offline stage extracts the basic geometric features. These are used during the online stage to compute the camera pose from a monocular image. Thus, we can handle the usual textureless 3D models used in industrial applications. A self-supplied and robust markerless tracking system that combines an edge tracker, a point based tracker and a 3D particle filter has also been designed to continuously update the camera pose. Our framework incorporates an automatic path-planning module. During the offline stage, the assembly/disassembly sequence is automatically deduced from the 3D model geometry. This information is used to generate the disassembly instructions for workers.

This paper addresses the problem of camera tracking and 3D reconstruction from image sequences, i.e., the monocular SLAM problem. Traditionally, this problem is solved using non-linear minimization techniques that are very accurate but hardly used in real time. This work presents a highly parallelizable random sampling approach based on Monte Carlo simulations that fits very well on the graphics hardware. The proposed algorithm achieves the same precision as non linear optimization, getting real time performance running on commodity graphics hardware. Both accuracy and performance are evaluated using synthetic data and real video sequences captured with a hand-held camera. Moreover, results are compared with an implementation of Bundle Adjustment showing that the presented method gets similar results in much less time.

Volume rendering has been widely used in different fields where several rendering algorithms have been developed, such as shear-warp, ray casting or splatting. But independently of the rendering method, transfer functions are usually used for mapping values and other properties of the volume into colors. As an improvement of transfer functions, style transfer functions are being used, where sphere maps extracted from artwork are used instead of plain colors. In this paper, we propose an interactive designer that would allow the user to create styles in an easy way, and shade them with just a color or a texture. In addition, it guarantees a coherent illumination, making it possible to easily use style transfer functions to achieve realistic rendering.

Currently, many visualization methods are used in computer assisted medicine. It is commonly considered that a unique visualization scheme makes difficult the interaction and limits the quality and quantity of the information shown. In this paper we study the specific requirements of a maxillofacial surgery simulation tool for facial appearance prediction. The different stages of the application lead to present medical information in different ways. We propose to adapt visualization techniques to give a more suitable answer to these needs: a hybrid volumetric and polygonal visualization for the planning stage, as well as a scheme for surgery definition in 2D. Finally, we propose the use of mesh visualization for the simulated model, which previously requires the 3D reconstruction of the surface in order to be visualized.

Abstract A GPU capable method for surface reconstruction from unorganized point clouds without additional information, called GLT (GPU Local Triangulation), is presented. The main objective of this research is the generation of a GPU interpolating reconstruction based on local Delaunay triangulations, inspired by a pre‐existing reconstruction algorithm. Current graphics hardware accelerated algorithms are approximating approaches, where the final triangulation is usually performed through either marching cubes or marching tetrahedras. GPU‐compatible methods and data structures to perform normal estimation and the local triangulation have been developed, plus a variation of the Bitonic Merge Sort algorithm to work with multi‐lists. Our method shows an average gain of one order of magnitude over previous research.

Maintenance operations have a great impact in the security and life expectancy of any product. Virtual Reality can help reducing costs and design time by moving testing from physical mock-ups to virtual ones. VR allows the application of additional tools not possible in a physical mock-up. The increase in power of current computers allows their use not only for visualization and interaction but also for geometrical analysis and reasoning. These capabilities provide designers with tools that can increase their productivity. This paper proposes methods for the generation of selective disassembly plans and the generation of collision free trajectories in the context of interactive simulation of maintenance operations. These methods provide a design team with additional tools to asses a design, prior to the construction of physical mock-ups and, therefore, reducing the cost of a project

The collision problem appears within many fields. The specific characteristics that can be identified in different problems lead to the broad set of specialized algorithms that appear in the literature. This paper deals with the first step needed to address the collision problem taxonomy challenge: a survey that compiles and suggests a set of characteristics that could be used to discriminate collision problems, i.e. to generate the taxonomy.

Abstract This paper describes a Collision Method for massive virtual environments composed of millions of triangles. It has been applied in the aeronautics industry for maintainability simulations using virtual aircraft engine mock‐ups. The method performs well and has a good interactive frame rate even when it is used for computing force feedback with haptic devices. Space sorting problems chiefly related to voxel techniques, such as memory requirements and optimal voxel size, have been solved. We use advanced memory structures and hashing techniques. To find the optimal voxel size, several analytical solutions have been proposed and compared. These solutions are based on the performance cost function of the algorithm used. Experiments have been undertaken to verify these analytical solutions.

This paper describes a haptic system for maintainability simulation in aeronautics, called REVIMA (Virtual Reality for Maintainability). In this project a software-hardware tool is designed and built to realistically simulate assembly-disassembly operations. It also helps to perform accessibility, interference and maintainability analysis by using virtual reality techniques without physical mock-ups. The system gives the user a reliable and realistic response. In order to achieve these requirements, the device has a workspace similar to the size of a turbo-engine. In addition this workspace can be placed in different positions to study ergonomics aspects of the simulated tasks.