<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Projects | Carlos Hernández Matas, PhD</title>
    <link>https://carlos.hernandez.im/project/</link>
      <atom:link href="https://carlos.hernandez.im/project/index.xml" rel="self" type="application/rss+xml" />
    <description>Projects</description>
    <generator>Source Themes Academic (https://sourcethemes.com/academic/)</generator><language>en-us</language><copyright>© 2024</copyright><lastBuildDate>Thu, 01 Aug 2024 00:00:00 +0000</lastBuildDate>
    <image>
      <url>img/map[gravatar:%!s(bool=false) shape:circle]</url>
      <title>Projects</title>
      <link>https://carlos.hernandez.im/project/</link>
    </image>
    
    <item>
      <title>TomTom ADAS &amp; ADS Maps (2024 - Now)</title>
      <link>https://carlos.hernandez.im/project/tomtom/</link>
      <pubDate>Thu, 01 Aug 2024 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/tomtom/</guid>
      <description>&lt;p&gt;At 
&lt;a href=&#34;https://www.tomtom.com&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;TomTom&lt;/a&gt; I&amp;rsquo;m the Engineering Manager of the Road Markings team, where I lead a talented team of engineers and applied scientists. This team building automatic pipelines for detecting and categorizing all kinds of signs painted on the roads. Think about lane separators, pedrestrian crosses, surface signs, etc.&lt;/p&gt;
&lt;p&gt;This extracted information is then used for building maps to be used on Advanced Driver-Assistance System (ADAS) and Advanced Driving Systems (ADS) for Autonomous Vehicles, spearheading the future of autonomous driving.&lt;/p&gt;
</description>
    </item>
    
    <item>
      <title>Oivi (2021 - 2023)</title>
      <link>https://carlos.hernandez.im/project/oivi/</link>
      <pubDate>Sun, 07 Feb 2021 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/oivi/</guid>
      <description>&lt;p&gt;
&lt;a href=&#34;https://www.oivi.co&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;Oivi&lt;/a&gt; is a company developing an integrated system for perfoming accessible screening allowing for early detection of diabetic retinopathy.&lt;/p&gt;
&lt;p&gt;My involvement consisted on owning and managing, from design to implementation, the retinal image processing, registration and fusion pipelines for standard and challenging sets of images. I also handled the Intellectual Property Rights (IPR) management and I was lead in international grant applications.&lt;/p&gt;
</description>
    </item>
    
    <item>
      <title>Aeolus Robot (2017 - 2021)</title>
      <link>https://carlos.hernandez.im/project/aeolus-robot/</link>
      <pubDate>Fri, 30 Jun 2017 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/aeolus-robot/</guid>
      <description>&lt;p&gt;
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/0ceVWiIH00g&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
  &lt;/p&gt;
&lt;p&gt;
&lt;a href=&#34;https://aeolusbot.com/&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;Aeolus Robotics&lt;/a&gt; is a service robotics startup, currently focused on providing assistive robots for elderly care facilities. I joined the startup in its early stages. I own features from specification to delivery and coordinate across teams to ensure agile development. I also perform algorithm development and testing, as well as hardware-software integration, for robot perception, grasping, motion and reactive planning. I have also led tech demos for investors, as well as knowledge transfer workshops for other developers on how to work with the robots. Additionally, I have performed technology scouting duties at international robotic conferences.&lt;/p&gt;
&lt;div class=&#34;container&#34;&gt;
  &lt;div class=&#34;row&#34;&gt;
    &lt;div class=&#34;col-12 col-md-6&#34;&gt;
      
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/0djXBwG1b0c&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
 &amp;nbsp;
    &lt;/div&gt;
    &lt;div class=&#34;col-12 col-md-6&#34;&gt;
      
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/LmkHKTxyOqU&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
 &amp;nbsp;
    &lt;/div&gt;
  &lt;/div&gt;
  &lt;div class=&#34;row&#34;&gt;
    &lt;div class=&#34;col-12 col-md-6&#34;&gt;
      
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/t12Efhy5L3Q?start=361&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
 &amp;nbsp;
    &lt;/div&gt;
    &lt;div class=&#34;col-12 col-md-6&#34;&gt;
      
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/mx_sIwiMcwI?start=85&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
 &amp;nbsp;
    &lt;/div&gt;
  &lt;/div&gt;
&lt;/div&gt;
&lt;p&gt;The videos above contain some demonstrations done recently on TV and at a client facility. The videos below shows part of a demo we did with our initial prototype at CES 2018&lt;/p&gt;
&lt;div class=&#34;container&#34;&gt;
  &lt;div class=&#34;row&#34;&gt;
    &lt;div class=&#34;col-12 col-md-6&#34;&gt;
      
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/xgZW6bfSEiw&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
 &amp;nbsp;
    &lt;/div&gt;
    &lt;div class=&#34;col-12 col-md-6&#34;&gt;
      
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/B8yQmPhQPBs&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
 &amp;nbsp;
    &lt;/div&gt;
  &lt;/div&gt;
&lt;/div&gt;
</description>
    </item>
    
    <item>
      <title>REVAMMAD (2013 - 2017)</title>
      <link>https://carlos.hernandez.im/project/revammad/</link>
      <pubDate>Sun, 15 Sep 2013 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/revammad/</guid>
      <description>&lt;p&gt;
&lt;a href=&#34;https://cordis.europa.eu/project/id/316990&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;REVAMMAD (REtinal Vascular Modelling, Measurement and Diagnosis)&lt;/a&gt; was a European Union project aimed at combatting some of the EU&amp;rsquo;s most prevalent chronic medical conditions using retinal imaging. The 3.8 million euro project trained a new generation of interdisciplinary scientists for the academic, clinical and industrial sectors, and triggered a new wave of biomedical interventions. The PhD students received training by some of the EU&amp;rsquo;s leading academics and practitioners to achieve further advances in diagnosis, prognosis and prevention of diseases such as diabetes, hypertension, stroke and coronary heart disease and retinal diseases.&lt;/p&gt;
&lt;p&gt;My involvement in this project mainly consisted on employing diverse computer vision techniques, such as image registration or super resolution, to improve image quality as well as facilitating the analysis of retinal images by clinicians, allowing them to perform faster and more accurate diagnoses. This work was performed at the Institute of Computer Science of the Foundation for 
&lt;a href=&#34;https://ics.forth.gr/&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;Research and Technology - Hellas (FORTH-ICS)&lt;/a&gt; lead to a Ph.D. from the 
&lt;a href=&#34;https://www.uoc.gr/en/&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;University of Crete&lt;/a&gt;, the release of the software created as well as a dataset that is becoming the &lt;em&gt;de facto&lt;/em&gt; standard for retinal image registration evaluation, and a series of publications that are listed below.&lt;/p&gt;
</description>
    </item>
    
    <item>
      <title>Color Blind Simulator (2011 - Now)</title>
      <link>https://carlos.hernandez.im/project/color-blind-simulator/</link>
      <pubDate>Sun, 27 Feb 2011 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/color-blind-simulator/</guid>
      <description>&lt;p&gt;
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/KcB_NhUNP0I&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;

 &lt;/p&gt;
&lt;p&gt;To help raise awareness on color blindness, I made a program that retrieves images from a live video feed or from a video file stored in a drive and performs a color transformation to show how the video would be seen by one of the three kinds of dichromat color blind people: people with protanopia, deuteranopia and tritanopia. Initial versions of this project were developed on C++, but the current implementation runs on Python 3.&lt;/p&gt;
&lt;p&gt;If you enjoyed this work, please consider buying me a coffee in the link below&lt;/p&gt;
&lt;script type=&#39;text/javascript&#39; src=&#39;https://ko-fi.com/widgets/widget_2.js&#39;&gt;&lt;/script&gt;&lt;script type=&#39;text/javascript&#39;&gt;kofiwidget2.init(&#39;Support Me on Ko-fi&#39;, &#39;#29abe0&#39;, &#39;Y8Y31HKCP&#39;);kofiwidget2.draw();&lt;/script&gt;
&lt;p&gt; &lt;/p&gt;
&lt;p&gt;If you are interested in having your video converted to one of these simulations, you can reach out to me 
&lt;a href=&#34;https://carlos.hernandez.im/#contact&#34;&gt;in the contact section&lt;/a&gt;.&lt;/p&gt;
&lt;p&gt;
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/-UNQxXfY2AI&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;

 

&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/ZW2NVBvI9TA&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
&lt;/p&gt;
</description>
    </item>
    
    <item>
      <title>Active 3D Segmentation (2011)</title>
      <link>https://carlos.hernandez.im/project/active-3d-segmentation/</link>
      <pubDate>Sat, 15 Jan 2011 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/active-3d-segmentation/</guid>
      <description>&lt;p&gt;This project was developed as a Master Thesis for the 
&lt;a href=&#34;https://www.kth.se/en&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;KTH Royal Institute of Technology&lt;/a&gt;. It consists on a study about the capabilities of applying the depth maps generated by Kinect to image segmentation. Specifically, it focuses on providing an alternative attention mechanism to select fixations points for the segmentation, as well as two sets of cues for the object-background separation. It allows to be able to automatically detect objects of interest on a scene, without requiring manual input.&lt;/p&gt;
&lt;p&gt;Developed in C++.&lt;/p&gt;

&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/NrdCriH6xIA&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;

</description>
    </item>
    
    <item>
      <title>Bumpy Robot (2010)</title>
      <link>https://carlos.hernandez.im/project/bumpy-robot/</link>
      <pubDate>Thu, 30 Dec 2010 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/bumpy-robot/</guid>
      <description>&lt;p&gt;
&lt;div style=&#34;position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;&#34;&gt;
  &lt;iframe src=&#34;https://www.youtube.com/embed/VZnamTVljyg&#34; style=&#34;position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;&#34; allowfullscreen title=&#34;YouTube Video&#34;&gt;&lt;/iframe&gt;
&lt;/div&gt;
  &lt;/p&gt;
&lt;p&gt;Bumpy was a robot developed from scratch for the Fall 2010 intake of the 
&lt;a href=&#34;https://www.kth.se/student/kurser/kurs/DD2425?l=en&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;DD2425 Robotics and Autonomous Systems&lt;/a&gt; course from the 
&lt;a href=&#34;https://www.kth.se/en&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;KTH Royal Institute of Technology&lt;/a&gt;. Bumpy was able to performing the whole set of tasks for the course:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Detecting victims and bombs with their distances to the robot, and decoding barcodes&lt;/li&gt;
&lt;li&gt;Rescuing victims in order.&lt;/li&gt;
&lt;li&gt;Avoiding bombs.&lt;/li&gt;
&lt;li&gt;If a bomb and a victim are in the same cell, and the victim is closer, rescue it.&lt;/li&gt;
&lt;li&gt;Usage of the map (example: if we rescue victim 3 and we already know where 4 is, we go straight to pick it up).&lt;/li&gt;
&lt;li&gt;Graphical display of the map&lt;/li&gt;
&lt;li&gt;Recovering from impacts against walls.&lt;/li&gt;
&lt;li&gt;Entering the maze from an outside point.&lt;/li&gt;
&lt;li&gt;Exiting the maze with the rescued victims (using map navigation).&lt;/li&gt;
&lt;/ul&gt;
</description>
    </item>
    
    <item>
      <title>Sound Ray Tracer (2008)</title>
      <link>https://carlos.hernandez.im/project/sound-ray-tracer/</link>
      <pubDate>Tue, 15 Jan 2008 00:00:00 +0000</pubDate>
      <guid>https://carlos.hernandez.im/project/sound-ray-tracer/</guid>
      <description>&lt;p&gt;This project was developed in collaboration with Diana Gómez Olmedilla as a Bachelor Thesis for the 
&lt;a href=&#34;https://www.bth.se/eng/&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;BTH Blekinge Institute of Technology&lt;/a&gt; and the 
&lt;a href=&#34;https://www.upm.es/internacional&#34; target=&#34;_blank&#34; rel=&#34;noopener&#34;&gt;UPM Technical University of Madrid&lt;/a&gt;.&lt;/p&gt;
&lt;p&gt;Given the size of an orthogonal room, the reflection coefficient of its walls, the location of a source of sound and the location of a listening device, this program calculates the normalized impulse response between those 2 points by taking into account all the reflections of the sound in the walls.&lt;/p&gt;
&lt;p&gt;Developed in MATLAB, the code is in the second half of the thesis report, which can be found in the publication below.&lt;/p&gt;
</description>
    </item>
    
  </channel>
</rss>
