Open Access
Accurate pupil center detection in off-the-shelf eye tracking systems using convolutional neural networks
(MDPI, 2021) Larumbe Bergera, Andoni; Garde Lecumberri, Gonzalo; Porta Cuéllar, Sonia; Cabeza Laguna, Rafael; Villanueva Larre, Arantxa; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren; Universidad Pública de Navarra / Nafarroako Unibertsitate Publikoa
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicability in many research areas. In this paper, a video-oculography method based on convolutional neural networks (CNNs) for pupil center detection over webcam images is proposed. As the first contribution of this work and in order to train the model, a pupil center manual labeling procedure of a facial landmark dataset has been performed. The model has been tested over both real and synthetic databases and outperforms state-of-the-art methods, achieving pupil center estimation errors below the size of a constricted pupil in more than 95% of the images, while reducing computing time by a 8 factor. Results show the importance of use high quality training data and well-known architectures to achieve an outstanding performance.
Open Access
U2Eyes: a binocular dataset for eye tracking and gaze estimation
(IEEE, 2019) Porta Cuéllar, Sonia; Bossavit, Benoît; Cabeza Laguna, Rafael; Larumbe Bergera, Andoni; Garde Lecumberri, Gonzalo; Villanueva Larre, Arantxa; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren
Theory shows that huge amount of labelled data are needed in order to achieve reliable classification/regression methods when using deep/machine learning techniques. However, in the eye tracking field, manual annotation is not a feasible option due to the wide variability to be covered. Hence, techniques devoted to synthesizing images show up as an opportunity to provide vast amounts of annotated data. Considering that the well-known UnityEyes tool provides a framework to generate single eye images and taking into account that both eyes information can contribute to improve gaze estimation accuracy we present U2Eyes dataset, that is publicly available. It comprehends about 6 million of synthetic images containing binocular data. Furthermore, the physiology of the eye model employed is improved, simplified dynamics of binocular vision are incorporated and more detailed 2D and 3D labelled data are provided. Additionally, an example of application of the dataset is shown as work in progress. Employing U2Eyes as training framework Supervised Descent Method (SDM) is used for eyelids segmentation. The model obtained as result of the training process is then applied on real images from GI4E dataset showing promising results.
Open Access
SeTA: semiautomatic tool for annotation of eye tracking images
(Association for Computing Machinery, 2019) Larumbe Bergera, Andoni; Porta Cuéllar, Sonia; Cabeza Laguna, Rafael; Villanueva Larre, Arantxa; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren
Availability of large scale tagged datasets is a must in the field of deep learning applied to the eye tracking challenge. In this paper, the potential of Supervised-Descent-Method (SDM) as a semiautomatic labelling tool for eye tracking images is shown. The objective of the paper is to evidence how the human effort needed for manually labelling large eye tracking datasets can be radically reduced by the use of cascaded regressors. Different applications are provided in the fields of high and low resolution systems. An iris/pupil center labelling is shown as example for low resolution images while a pupil contour points detection is demonstrated in high resolution. In both cases manual annotation requirements are drastically reduced.
Open Access
Relevance of sex, age and gait kinematics when predicting fall-risk and mortality in older adults
(Elsevier, 2020) Porta Cuéllar, Sonia; Martínez Ramírez, Alicia; Millor Muruzábal, Nora; Gómez Fernández, Marisol; Izquierdo Redín, Mikel; Ingeniería Eléctrica, Electrónica y de Comunicación; Estadística, Informática y Matemáticas; Ciencias de la Salud; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren; Estatistika, Informatika eta Matematika; Osasun Zientziak; Gobierno de Navarra / Nafarroako Gobernua, 87/10
Approximately one-third of elderly people fall each year with severe consequences, including death. The aim of this study was to identify the most relevant features to be considered to maximize the accuracy of a logistic regression model designed for prediction of fall/mortality risk among older people. This study included 261 adults, aged over 65 years. Men and women were analyzed separately because sex stratification was revealed as being essential for our purposes of feature ranking and selection. Participants completed a 3-m walk test at their own gait velocity. An inertial sensor attached to their lumbar spine was used to record acceleration data in the three spatial directions. Signal processing techniques allowed the extraction of 21 features representative of gait kinematics, to be used as predictors to train and test the model. Age and gait speed data were also considered as predictors. A set of 23 features was considered. These features demonstrate to be more or less relevant depending on the sex of the cohort under analysis and the classification label (risk of falls and mortality). In each case, the minimum size subset of relevant features is provided to show the maximum accuracy prediction capability. Gait speed has been largely used as the single feature for the prediction fall risk among older adults. Nevertheless, prediction accuracy can be substantially improved, reaching 70% in some cases, if the task of training and testing the model takes into account some other features, namely, sex, age and gait kinematic parameters. Therefore we recommend considering sex, age and step regularity to predict fall-risk.
Open Access
Exact inter-discharge interval distribution of motor unit firing patterns with gamma model
(Springer, 2019) Navallas Irujo, Javier; Porta Cuéllar, Sonia; Malanda Trigueros, Armando; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren
Inter-discharge interval distribution modeling of the motor unit firing pattern plays an important role in electromyographic decomposition and the statistical analysis of firing patterns. When modeling firing patterns obtained from automatic procedures, false positives and false negatives can be taken into account to enhance performance in estimating firing pattern statistics. Available models of this type, however, are only approximate and use Gaussian distributions, which are not strictly suitable for modeling renewal point processes. In this paper, the theory of point processes is used to derive an exact solution to the distribution when a gamma distribution is used to model the physiological firing pattern. Besides being exact, the solution provides a way to model the skewness of the inter-discharge distribution, and this may make it possible to obtain a better fit with available experimental data. In order to demonstrate potential applications of the model, we use it to obtain a maximum likelihood estimator of firing pattern statistics. Our tests found this estimator to be reliable over a wide range of firing conditions, whether dealing with real or simulated firing patterns, the proposed solution had better agreement than other models.
Open Access
Introducing I2Head database
(Association for Computing Machinery, 2018) Martinikorena Aranburu, Ion; Cabeza Laguna, Rafael; Villanueva Larre, Arantxa; Porta Cuéllar, Sonia; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren
I2Head database has been created with the aim to become an optimal reference for low cost gaze estimation. It exhibits the following outstanding characteristics: it takes into account key aspects of low resolution eye tracking technology; it combines images of users gazing at different grids of points from alternative positions with registers of user's head position and it provides calibration information of the camera and a simple 3D head model for each user. Hardware used to build the database includes a 6D magnetic sensor and a webcam. A careful calibration method between the sensor and the camera has been developed to guarantee the accuracy of the data. Different sessions have been recorded for each user including not only static head scenarios but also controlled displacements and even free head movements. The database is an outstanding framework to test both gaze estimation algorithms and head pose estimation methods.
Open Access
Synthetic gaze data augmentation for improved user calibration
(Springer, 2021) Garde Lecumberri, Gonzalo; Larumbe Bergera, Andoni; Porta Cuéllar, Sonia; Cabeza Laguna, Rafael; Villanueva Larre, Arantxa; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren; Institute of Smart Cities - ISC; Ingeniería Eléctrica, Electrónica y de Comunicación
In this paper, we focus on the calibration possibilitiesó of a deep learning based gaze estimation process applying transfer learning, comparing its performance when using a general dataset versus when using a gaze specific dataset in the pretrained model. Subject calibration has demonstrated to improve gaze accuracy in high performance eye trackers. Hence, we wonder about the potential of a deep learning gaze estimation model for subject calibration employing fine-tuning procedures. A pretrained Resnet-18 network, which has great performance in many computer vision tasks, is fine-tuned using user’s specific data in a few shot adaptive gaze estimation approach. We study the impact of pretraining a model with a synthetic dataset, U2Eyes, before addressing the gaze estimation calibration in a real dataset, I2Head. The results of the work show that the success of the individual calibration largely depends on the balance between fine-tuning and the standard supervised learning procedures and that using a gaze specific dataset to pretrain the model improves the accuracy when few images are available for calibration. This paper shows that calibration is feasible in low resolution scenarios providing outstanding accuracies below 1.5 ∘ ∘ of error.
Open Access
Low cost gaze estimation: knowledge-based solutions
(IEEE, 2020) Martinikorena Aranburu, Ion; Larumbe Bergera, Andoni; Ariz Galilea, Mikel; Porta Cuéllar, Sonia; Cabeza Laguna, Rafael; Villanueva Larre, Arantxa; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren
Eye tracking technology in low resolution scenarios is not a completely solved issue to date. The possibility of using eye tracking in a mobile gadget is a challenging objective that would permit to spread this technology to non-explored fields. In this paper, a knowledge based approach is presented to solve gaze estimation in low resolution settings. The understanding of the high resolution paradigm permits to propose alternative models to solve gaze estimation. In this manner, three models are presented: a geometrical model, an interpolation model and a compound model, as solutions for gaze estimation for remote low resolution systems. Since this work considers head position essential to improve gaze accuracy, a method for head pose estimation is also proposed. The methods are validated in an optimal framework, I2Head database, which combines head and gaze data. The experimental validation of the models demonstrates their sensitivity to image processing inaccuracies, critical in the case of the geometrical model. Static and extreme movement scenarios are analyzed showing the higher robustness of compound and geometrical models in the presence of user’s displacement. Accuracy values of about 3◦ have been obtained, increasing to values close to 5◦ in extreme displacement settings, results fully comparable with the state-of-the-art.
Open Access
Sliding window averaging in normal and pathological motor unit action potential trains
(Elsevier, 2018) Malanda Trigueros, Armando; Navallas Irujo, Javier; Rodríguez Falces, Javier; Porta Cuéllar, Sonia; Fernández Martínez, Miguel; Ingeniería Eléctrica, Electrónica y de Comunicación; Ingeniaritza Elektrikoa, Elektronikoaren eta Telekomunikazio Ingeniaritzaren
Objective: To evaluate the performance of a recently proposed motor unit action potential (MUAP) averaging method based on a sliding window, and compare it with relevant published methods in normal and pathological muscles. Methods: Three versions of the method (with different window lengths) were compared to three relevant published methods in terms of signal analysis-based merit figures and MUAP waveform parameters used in the clinical practice. 218 MUAP trains recorded from normal, myopathic, subacute neurogenic and chronic neurogenic muscles were analysed. Percentage scores of the cases in which the methods obtained the best performance or a performance not significantly worse than the best were computed. Results: For signal processing figures of merit, the three versions of the new method performed better (with scores of 100, 86.6 and 66.7%) than the other three methods (66.7, 25 and 0%, respectively). In terms of MUAP waveform parameters, the new method also performed better (100, 95.8 and 91.7%) than the other methods (83.3, 37.5 and 25%). Conclusions: For the types of normal and pathological muscle studied, the sliding window approach extracted more accurate and reliable MUAP curves than other existing methods. Significance: The new method can be of service in quantitative EMG.
Open Access
Motor unit profile: a new way to describe the scanning-EMG potential
(Elsevier, 2017) Corera Orzanco, Íñigo; Malanda Trigueros, Armando; Rodríguez Falces, Javier; Porta Cuéllar, Sonia; Navallas Irujo, Javier; Ingeniería Eléctrica y Electrónica; Ingeniaritza Elektrikoa eta Elektronikoa
The motor unit profile, a representation of the trajectories of positive and negative turns of a scanning-EMG signal, is a new way to characterize the motor unit potential. Such characterization allows quantification of the scanning-EMG signal's complexity, which is closely related to the anatomy and physiology of the motor unit. To extract the motor unit profile, an algorithm that detects the turns of the scanning-EMG signal and links them using point-tracking techniques has been developed. The performance of this algorithm is sensitive to three parameters: the turn detection threshold, the maximum tracking interval threshold, and the trajectory purge threshold. Real scanning-EMG signals have been used to analyze the algorithm's behavior and the influence of the algorithm's parameters and to determine which parameter values provide the best performance.