Victor Villena-Martinez

A Quantitative Comparison of Calibration Methods for RGB-D Sensors Using Different Technologies

RGB-D (Red Green Blue and Depth) sensors are devices that can provide color and depth information from a scene at the same time. Recently, they have been widely used in many solutions due to their commercial growth from the entertainment market to many diverse areas (e.g., robotics, CAD, etc.). In the research community, these devices have had good uptake due to their acceptable level of accuracy for many applications and their low cost, but in some cases, they work at the limit of their sensitivity, near to the minimum feature size that can be perceived. For this reason, calibration processes are critical in order to increase their accuracy and enable them to meet the requirements of such kinds of applications. To the best of our knowledge, there is not a comparative study of calibration algorithms evaluating its results in multiple RGB-D sensors. Specifically, in this paper, a comparison of the three most used calibration methods have been applied to three different RGB-D sensors based on structured light and time-of-flight. The comparison of methods has been carried out by a set of experiments to evaluate the accuracy of depth measurements. Additionally, an object reconstruction application has been used as example of an application for which the sensor works at the limit of its sensitivity. The obtained results of reconstruction have been evaluated through visual inspection and quantitative measurements.

A long short-term memory based Schaeffer gesture recognition system

In this work, a Schaeffer language recognition system is proposed in order to help autistic children overcome communicative disorders. Using Schaeffer language as a speech and language therapy, improves children communication skills and at the same time the understanding of language productions. Nevertheless, the teaching process of children in performing gestures properly is not straightforward. For this purpose, this system will teach children with autism disorder the correct way to communicate using gestures in combination with speech reproduction. The main purpose is to accelerate the learning process and increase children interest by using a technological approach. Several recurrent neural network-based approaches have been tested, such as vanilla recurrent neural networks, long short-term memory networks,and gated recurrent unit-based models. In order to select the most suitable model, an extensive comparison has been conducted reporting a 93.13% classification success rate over a subset of 25 Schaeffer gestures by using an long short-term memory-based approach. Our dataset consists on pose-based features such as angles and euclidean distances extracted from the raw skeletal data provided by a Kinect v2 sensor.

3D non-rigid registration using color: Color Coherent Point Drift

Abstract Research into object deformations using computer vision techniques has been under intense study in recent years. A widely used technique is 3D non-rigid registration to estimate the transformation between two instances of a deforming structure. Despite many previous developments on this topic, it remains a challenging problem. In this paper we propose a novel approach to non-rigid registration combining two data spaces in order to robustly calculate the correspondences and transformation between two data sets. In particular, we use point color as well as 3D location as these are the common outputs of RGB-D cameras. We have propose the Color Coherent Point Drift (CCPD) algorithm (an extension of the CPD method (Myronenko and Song, 2010)). Evaluation is performed using synthetic and real data. The synthetic data includes easy shapes that allow evaluation of the effect of noise, outliers and missing data. Moreover, an evaluation of realistic figures obtained using Blensor is carried out. Real data acquired using a general purpose Primesense Carmine sensor is used to validate the CCPD for real shapes. For all tests, the proposed method is compared to the original CPD showing better results in registration accuracy in most cases.

3D Body Registration from RGB-D Data with Unconstrained Movements and Single Sensor

In this paper, the problem of 3D body registration using a single RGB-D sensor is approached. It has been guided by three main requirements: low-cost, unconstrained movement and accuracy. In order to fit them, an iterative registration method for accurately aligning data from single RGB-D sensor is proposed. The data is acquired while a person rotates in front of the camera, without the need of any external marker or constraint about its pose. The articulated alignment is carried out in a model-free approach in order to be more consistent with the real data. The iterative method is divided in stages, contributing to each other by the refinement of a specific part of the acquired data. The exploratory results validate the proposed method that is able to feed on itself in each iteration improving the final result by a progressive iteration, with the required precision under the conditions of affordability and unconstrained movement acquisition.

An Iterative Method for 3D Body Registration Using a Single RGB-D Sensor

In this paper, the problem of 3D body registration using a single RGB-D sensor is approached. It has been guided by three main requirements: low-cost, unconstrained movement and accuracy. In order to fit them, an iterative registration method for accurately aligning data from single RGB-D sensor is proposed. The data is acquired while a person rotates in front of the camera, without the need of any external marker or constraint about its pose. The articulated alignment is carried out in a model-free approach in order to be more consistent with the real data. The iterative method is divided in stages, contributing to each other by the refinement of a specific part of the acquired data. The exploratory results validate the proposed method that is able to feed on itself in each iteration improving the final result by a progressive iteration, with the required precision under the conditions of affordability and unconstrained movement acquisition.

Topology Preserving Self-Organizing Map of Features in Image Space for Trajectory Classification

Self-Organizing maps (SOM) are able to preserve topological information in the projecting space. Structure and learning algorithm of SOMs restrict the topological preservation in the map. Adjacent neurons share similar vector features. However, topological preservation from the input space is not always accomplished. In this paper, we propose a novel self-organizing feature map that is able to preserve the topological information about the scene in the image space. Extracted features in adjacent areas of an image are explicitly in adjacent areas of the self-organizing map preserving input topology (SOM-PINT). The SOM-PINT has been applied to represent and classify trajectories into high level of semantic understanding from video sequences. Experiments have been carried out using the Shopping Centre dataset of the CAVIAR database taken into account the global behaviour of an individual. Results confirm the input preservation topology in image space to obtain high performance classification for trajectory classification in contrast of traditional SOM.