GPGPU implementation of growing neural gas: application to 3D scene reconstruction

Self-organising neural models have the ability to provide a good representation of the input space. In particular the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time-consuming, especially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This paper proposes a Graphics Processing Unit (GPU) parallel implementation of the GNG with Compute Unified Device Architecture (CUDA). In contrast to existing algorithms, the proposed GPU implementation allows the acceleration of the learning process keeping a good quality of representation. Comparative experiments using iterative, parallel and hybrid implementations are carried out to demonstrate the effectiveness of CUDA implementation. The results show that GNG learning with the proposed implementation achieves a speed-up of 6× compared with the single-threaded CPU implementation. GPU implementation has also been applied to a real application with time constraints: acceleration of 3D scene reconstruction for egomotion, in order to validate the proposal.

3D Feature Extraction and Modelling for SLAM

Paper submitted to the 39th International Symposium on Robotics ISR 2008, Seoul, South Korea, October 15-17, 2008.

Portable 3D laser-camera calibration system with color fusion for SLAM

Nowadays, the use of RGB-D sensors have focused a lot of research in computer vision and robotics. These kinds of sensors, like Kinect, allow to obtain 3D data together with color information. However, their working range is limited to less than 10 meters, making them useless in some robotics applications, like outdoor mapping. In these environments, 3D lasers, working in ranges of 20-80 meters, are better. But 3D lasers do not usually provide color information. A simple 2D camera can be used to provide color information to the point cloud, but a calibration process between camera and laser must be done. In this paper we present a portable calibration system to calibrate any traditional camera with a 3D laser in order to assign color information to the 3D points obtained. Thus, we can use laser precision and simultaneously make use of color information. Unlike other techniques that make use of a three-dimensional body of known dimensions in the calibration process, this system is highly portable because it makes use of small catadioptrics that can be placed in a simple manner in the environment. We use our calibration system in a 3D mapping system, including Simultaneous Location and Mapping (SLAM), in order to get a 3D colored map which can be used in different tasks. We show that an additional problem arises: 2D cameras information is different when lighting conditions change. So when we merge 3D point clouds from two different views, several points in a given neighborhood could have different color information. A new method for color fusion is presented, obtaining correct colored maps. The system will be tested by applying it to 3D reconstruction.

Portable 3D laser-camera calibration system with color fusion for SLAM

Paper submitted to the 43rd International Symposium on Robotics (ISR2012), Taipei, Taiwan, Aug. 29-31, 2012.

A parallel method for impulsive image noise removal on hybrid CPU/GPU systems

A parallel algorithm for image noise removal is proposed. The algorithm is based on peer group concept and uses a fuzzy metric. An optimization study on the use of the CUDA platform to remove impulsive noise using this algorithm is presented. Moreover, an implementation of the algorithm on multi-core platforms using OpenMP is presented. Performance is evaluated in terms of execution time and a comparison of the implementation parallelised in multi-core, GPUs and the combination of both is conducted. A performance analysis with large images is conducted in order to identify the amount of pixels to allocate in the CPU and GPU. The observed time shows that both devices must have work to do, leaving the most to the GPU. Results show that parallel implementations of denoising filters on GPUs and multi-cores are very advisable, and they open the door to use such algorithms for real-time processing.

Real-time 3D semi-local surface patch extraction using GPGPU

Feature vectors can be anything from simple surface normals to more complex feature descriptors. Feature extraction is important to solve various computer vision problems: e.g. registration, object recognition and scene understanding. Most of these techniques cannot be computed online due to their complexity and the context where they are applied. Therefore, computing these features in real-time for many points in the scene is impossible. In this work, a hardware-based implementation of 3D feature extraction and 3D object recognition is proposed to accelerate these methods and therefore the entire pipeline of RGBD based computer vision systems where such features are typically used. The use of a GPU as a general purpose processor can achieve considerable speed-ups compared with a CPU implementation. In this work, advantageous results are obtained using the GPU to accelerate the computation of a 3D descriptor based on the calculation of 3D semi-local surface patches of partial views. This allows descriptor computation at several points of a scene in real-time. Benefits of the accelerated descriptor have been demonstrated in object recognition tasks. Source code will be made publicly available as contribution to the Open Source Point Cloud Library.

A robust and fast method for 6DoF motion estimation from generalized 3D data

Nowadays, there is an increasing number of robotic applications that need to act in real three-dimensional (3D) scenarios. In this paper we present a new mobile robotics orientated 3D registration method that improves previous Iterative Closest Points based solutions both in speed and accuracy. As an initial step, we perform a low cost computational method to obtain descriptions for 3D scenes planar surfaces. Then, from these descriptions we apply a force system in order to compute accurately and efficiently a six degrees of freedom egomotion. We describe the basis of our approach and demonstrate its validity with several experiments using different kinds of 3D sensors and different 3D real environments.

Parallel Computational Intelligence-Based Multi-Camera Surveillance System

In this work, we present a multi-camera surveillance system based on the use of self-organizing neural networks to represent events on video. The system processes several tasks in parallel using GPUs (graphic processor units). It addresses multiple vision tasks at various levels, such as segmentation, representation or characterization, analysis and monitoring of the movement. These features allow the construction of a robust representation of the environment and interpret the behavior of mobile agents in the scene. It is also necessary to integrate the vision module into a global system that operates in a complex environment by receiving images from multiple acquisition devices at video frequency. Offering relevant information to higher level systems, monitoring and making decisions in real time, it must accomplish a set of requirements, such as: time constraints, high availability, robustness, high processing speed and re-configurability. We have built a system able to represent and analyze the motion in video acquired by a multi-camera network and to process multi-source data in parallel on a multi-GPU architecture.

Trajectory-Based Morphological Operators: A Model for Efficient Image Processing

Mathematical morphology has been an area of intensive research over the last few years. Although many remarkable advances have been achieved throughout these years, there is still a great interest in accelerating morphological operations in order for them to be implemented in real-time systems. In this work, we present a new model for computing mathematical morphology operations, the so-called morphological trajectory model (MTM), in which a morphological filter will be divided into a sequence of basic operations. Then, a trajectory-based morphological operation (such as dilation, and erosion) is defined as the set of points resulting from the ordered application of the instant basic operations. The MTM approach allows working with different structuring elements, such as disks, and from the experiments, it can be extracted that our method is independent of the structuring element size and can be easily applied to industrial systems and high-resolution images.

A Comparative Study of Registration Methods for RGB-D Video of Static Scenes

The use of RGB-D sensors for mapping and recognition tasks in robotics or, in general, for virtual reconstruction has increased in recent years. The key aspect of these kinds of sensors is that they provide both depth and color information using the same device. In this paper, we present a comparative analysis of the most important methods used in the literature for the registration of subsequent RGB-D video frames in static scenarios. The analysis begins by explaining the characteristics of the registration problem, dividing it into two representative applications: scene modeling and object reconstruction. Then, a detailed experimentation is carried out to determine the behavior of the different methods depending on the application. For both applications, we used standard datasets and a new one built for object reconstruction.

Combining visual features and Growing Neural Gas networks for robotic 3D SLAM

The use of 3D data in mobile robotics provides valuable information about the robot’s environment. Traditionally, stereo cameras have been used as a low-cost 3D sensor. However, the lack of precision and texture for some surfaces suggests that the use of other 3D sensors could be more suitable. In this work, we examine the use of two sensors: an infrared SR4000 and a Kinect camera. We use a combination of 3D data obtained by these cameras, along with features obtained from 2D images acquired from these cameras, using a Growing Neural Gas (GNG) network applied to the 3D data. The goal is to obtain a robust egomotion technique. The GNG network is used to reduce the camera error. To calculate the egomotion, we test two methods for 3D registration. One is based on an iterative closest points algorithm, and the other employs random sample consensus. Finally, a simultaneous localization and mapping method is applied to the complete sequence to reduce the global error. The error from each sensor and the mapping results from the proposed method are examined.

Parallel relaxed and extrapolated algorithms for computing PageRank

In this paper, parallel Relaxed and Extrapolated algorithms based on the Power method for accelerating the PageRank computation are presented. Different parallel implementations of the Power method and the proposed variants are analyzed using different data distribution strategies. The reported experiments show the behavior and effectiveness of the designed algorithms for realistic test data using either OpenMP, MPI or an hybrid OpenMP/MPI approach to exploit the benefits of shared memory inside the nodes of current SMP supercomputers.

Panorámica: serious games, gamification y mucho más

La relación entre los videojuegos y el mundo de la educación es y ha sido tormentosa, con amores y odios, con sus altibajos. Pero lo que es indudable es que los videojuegos son una realidad en el mundo actual y una potente industria. Y además, los juegos siempre han jugado un papel fundamental en la educación. Aunque su incorporación a la actividad académica no ha sido todo lo ágil que hubiera sido conveniente, los videojuegos ya forman parte de la universidad. En este artículo vamos a presentar algunas de las iniciativas que hemos llevado a cabo desde que en el año 2002 incorporamos los videojuegos en nuestras actividades académicas, tanto docentes como investigadoras. MadUniversity es un videojuego que dio lugar a varios proyectos final de carrera de la Ingeniería en Informática. Screaming Racers es un videojuego diseñado y desarrollado para ser utilizado como plataforma de experimentación de técnicas en inteligencia artificial basadas en la neuroevolución. The Conference Interpreter (CoIn) es un videojuego para la práctica del inglés desarrollado para apoyar una tesis doctoral. GameLearning es una colección de minijuegos conceptuales para la adquisición de habilidades directivas. ABPgame es la aplicación de la metodología basada en proyectos a varias asignaturas de las titulaciones de Ingeniería en Informática y del Grado en Ingeniería Multimedia que realizan un proyecto común: un videojuego. PLMan es un sistema gamificado que ayuda a desarrollar habilidades de pensamiento lógico, a través del lenguaje Prolog. Nuestro objetivo es mostrar la utilidad de los videojuegos y sus múltiples aplicaciones en el entorno universitario: como objetos de aprendizaje por medio de videojuegos educativos (serious games); como proyectos informáticos complejos para ser desarrollados por nuestros estudiantes; como entorno de experimentación para comprobar la validez de las investigaciones en inteligencia artificial; y finalmente como filosofía a aplicar al campo de la educación, lo que se ha etiquetado como gamificación.

A Model to Visualize Information in a Complex Streets’ Network

his paper discusses a process to graphically view and analyze information obtained from a network of urban streets, using an algorithm that establishes a ranking of importance of the nodes of the network itself. The basis of this process is to quantify the network information obtained by assigning numerical values to each node, representing numerically the information. These values are used to construct a data matrix that allows us to apply a classification algorithm of nodes in a network in order of importance. From this numerical ranking of the nodes, the process finish with the graphical visualization of the network. An example is shown to illustrate the whole process.

Geometric 3D point cloud compression

The use of 3D data in mobile robotics applications provides valuable information about the robot’s environment but usually the huge amount of 3D information is unmanageable by the robot storage and computing capabilities. A data compression is necessary to store and manage this information but preserving as much information as possible. In this paper, we propose a 3D lossy compression system based on plane extraction which represent the points of each scene plane as a Delaunay triangulation and a set of points/area information. The compression system can be customized to achieve different data compression or accuracy ratios. It also supports a color segmentation stage to preserve original scene color information and provides a realistic scene reconstruction. The design of the method provides a fast scene reconstruction useful for further visualization or processing tasks.