Javier Solar

Followers

Following

Public Views

Interests

Uploads

Papers by Javier Solar

Improvement of a face detection system by evolutionary multi-objective optimization

Proceedings - HIS 2005: Fifth International Conference on Hybrid Intelligent Systems, 2005

This paper presents the application of evolutionary multi-objective optimization (EMO) to the imp... more This paper presents the application of evolutionary multi-objective optimization (EMO) to the improvement of a face detection system. The face detection system is based on the boosted cascade system, and analyzes image positions on different scales in a three-stepprocedure. Based on threshold settings, the algorithm decides whether to continue with the test on a finer scale at the current position. Thus, the thresholds for all scales and stages have a major influence on the performance of the system, and become the subject of the evolutionary optimization according to three objectives: low false positive rate, high detection rate and low processing time. The used EMO is the extension of the Standard Genetic Algorithm to the EMO case by using Fuzzy Pareto Dominance as a meta-heuristic. The application of this EMO to the face detection system is explored and discussed using images from a standard face detection benchmark dataset. From the runtime analysis of the EMO it can be concluded that the algorithm reliably approaches the Pareto set of the problem.

Download

Characterizing objectionable image content (pornography and nude images) of specific web segments: Chile as a case study

Proceedings - Third Latin American Web Congress, LA-WEB 2005, 2005

In this paper we perform a characterization of the objectionable image content (pornography and n... more In this paper we perform a characterization of the objectionable image content (pornography and nude images) of the Chilean web (.cl domain). This characterization is based on a novel objectionable image filtering system that detects porno and nude images. In an automated process we examine all .cl websites, and download a large number of the available images. Then, we analyze the images' contents and we detect human-skin blobs. Afterwards, special features are derived from these skin blobs (size, orientation, position, solidity, eccentricity, etc.). Finally, a statistical classifier is employed for classifying the images in three classes: normal, porno and nude. In addition to these, we characterize the three classes by their faces content. We present statistics of use to anyone concerned with the objectionable image content of the web in Chile. We believe that the described methodology can be employed for carrying out similar studies in other segments of the web, and also for the automatic discovering of new pornographic sites.

Download

Multiclass adaboost and coupled classifiers for object detection

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2008

Building robust and fast multiclass object detection systems is a important goal of computer visi... more Building robust and fast multiclass object detection systems is a important goal of computer vision. In the present paper we extend the well-known work of Viola and Jones on boosted cascade classifiers to the multiclass case with the goal of building multiclass and multiview object detectors. We propose to use nested cascades of multiclass boosted classifiers and we introduce the concept of coupled components in multiclass classifiers. We evaluate the system by building several multiview face detectors, each one built to detect a different number of classes. Thus, we present results showing how well the system scales. Promising results are obtained in the BioID database, showing the potentiality of the proposed methods for building object detectors.

Download

Face recognition using thermal infrared images for Human-Robot Interaction applications: A comparative study

2009 6th Latin American Robotics Symposium (LARS 2009), 2009

Abstract The aim of this work is to carry out a comparative study of face-recognition methods for... more

On the image content of the Chilean Web

Proceedings of the IEEE/LEOS 3rd International Conference on Numerical Simulation of Semiconductor Optoelectronic Devices (IEEE Cat. No.03EX726)

In this paper we perform a study of the image contents of the Chilean web (.cl domain) using auto... more In this paper we perform a study of the image contents of the Chilean web (.cl domain) using automatic feature extraction, content-based analysis and face detection algorithms. In an automated process we examine all .cl websites and download a large number of the images available (approx. 83,000). Then we extract several visual features (color, texture, shape, etc.) and we perform face detection using novel algorithms. Using this process we semi-automatically characterize the image content of the web in Chile in terms of the detected faces and the visual features obtained automatically. We present statistics of use to anyone concerned with the image content of the web in Chile. Our study is the first one to use content-based tools to determine the image contents of the web.

Download

UChile RoadRunners 2008 Team Description Paper

RoboCup …, 2008

Abstract. Our robotics group has teams in three leagues: humanoid, SPL, and @home. Taking advanta... more

Download

Robotics-Centered Outreach Activities: An Integrated Approach

IEEE Transactions on Education, 2010

Illumination compensation and normalization for robust face recognition using discrete cosine transform in logarithm domain

IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics), 2006

This paper presents a novel illumination normalization approach for face recognition under varyin... more This paper presents a novel illumination normalization approach for face recognition under varying lighting conditions. In the proposed approach, a discrete cosine transform (DCT) is employed to compensate for illumination variations in the logarithm domain. Since illumination variations mainly lie in the low-frequency band, an appropriate number of DCT coefficients are truncated to minimize variations under different lighting conditions. Experimental results on the Yale B database and CMU PIE database show that the proposed approach improves the performance significantly for the face images with large illumination variations. Moreover, the advantage of our approach is that it does not require any modeling steps and can be easily implemented in a real-time face recognition system.

Download

Advances in Domestic Service Robots in the Real World

Journal of Intelligent & Robotic Systems, 2014

The use of robots in domestic environments has increased largely in the recent years. There are a... more The use of robots in domestic environments has increased largely in the recent years. There are already several millions robots used for basic household chores. It is expected that domestic service robots become more and more useful, autonomous, multipurpose and reliable, and that they will be able to interact with humans and objects in the real world in a natural way. This challenge poses a large number of unsolved problems across many scientific disciplines. As a first effort in order to highlight relevant research in this area, in 2012 a first special issue on domestic service robots was published [1]. The addressed topics were cooperative human-robot interaction, manipulation of domestic objects, multimodal human-robot interaction, human detection and identification in uncontrolled setups, and high-level cognition in domestic environments. The aim of this second special issue on domestic service robots is to present recent advances in the

Download

Domestic Service Robots in the Real World

Journal of Intelligent & Robotic Systems, 2011

Download

Animal Welfare in Australia: Politics and Policy

Press have been essential to the realisation of this project and have my gratitude for their atte... more

Download

Javier Solar

Uploads

Papers by Javier Solar

Log In