[an error occurred while processing this directive]<meta name="keywords" content="Department of Computer Science, publications, by type, Machine Learning Systems"></meta><meta name="description" content="Publications, by type,Machine Learning Systems, Department of Computer Science, Oxford"></meta><title>Department of Computer Science: Machine Learning Systems: Publications
</title><style type="text/css">
			  .btn-toolbar::after, .btn-toolbar::before
			  {
			    content: " ";
			    display: table;
			  }

			
			.btn-toolbar
			{
			    margin-bottom: 2em;
			}

			  .btn-toolbar::after, .btn-group::after
			  {
			    clear: both;
			  }


			.btn-toolbar .btn, .btn-toolbar .btn-group
			  {
			    float: left;
			}
			
			.btn-group
			  {
			    display: inline-block;
			    position: relative;
			    vertical-align: middle;
			}

			  .btn-group > .btn:first-child:not(:last-child)
			  {
			    border-bottom-right-radius: 0;
			    border-top-right-radius: 0;
			}
			
			.btn-group > .btn:first-child
			  {
			    margin-left: 0;
			}

			.btn-group > .btn
			  {
			    float: left;
			    position: relative;
			}

			  .btn-group > .btn:not(:first-child):not(:last-child)
			  {
			    border-radius: 0;
			}
			
			  .btn-group > .btn:last-child:not(:first-child)
			  {
			    border-bottom-left-radius: 0;
			    border-top-left-radius: 0;
			}
			
			.btn-group .btn + .btn, .btn-group .btn + .btn-group, .btn-group .btn-group + .btn, .btn-group .btn-group + .btn-group
			  {
			    margin-left: -2px;
			}
			</style>[an error occurred while processing this directive]<div class="content-width"><ol class="breadcrumb text-uppercase"><li><a href="/">Home</a></li><li><a href="/research/">Research</a></li><li><a href="/research/cyberphysical/">Systems</a></li><li><a href="/research/cyberphysical/activities.html">activities</a></li><li><a href="/activities/MachineLearningSystems/">Machine Learning Systems</a></li><li class="active">Publications</li></ol></div><!-- MOBILE NAVIGATION --><div class="content-width space-below mobile-menu-placeholder visible-xs visible-sm"></div><!-- COLUMNS --><div class="container-fluid content-width"><div class="row"><div class="col-xs-12 col-md-9 col-md-push-3" id="maincontent"><!-- CONTENT (Column 2) --><div class="page-title"><h1>Machine Learning Systems: Publications
</h1></div><div role="toolbar" class="btn-toolbar"><div class="btn-group"><a href="/activities/publications/date/MachineLearningSystems.html" aria-hidden="true" type="button" class="btn btn-default"><span class="glyphicon glyphicon-sort-by-attributes"></span> By date</a><a href="/activities/publications/title/MachineLearningSystems.html" aria-hidden="true" type="button" class="btn btn-default"><span class="glyphicon glyphicon-sort-by-attributes"></span> By title</a><a href="/activities/publications/type/MachineLearningSystems.html" aria-hidden="true" type="button" class="btn btn-oxford-blue"><span class="glyphicon glyphicon-sort-by-attributes"></span> By type</a><a href="/activities/publications/bibtex/MachineLearningSystems.html" aria-hidden="true" type="button" class="btn btn-default"><span class="glyphicon glyphicon-file"></span> BibTex</a></div></div><h2>
      Journal papers
      </h2><ul class="list-unbulleted list-unbulleted-spaced"><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[1] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            DeepTIO: A Deep Thermal&#x2212;Inertial Odometry with Visual Hallucination </strong> </p>
	    <p class="publicationAuthors"> <em> M.R.U. Saputra P.P.B. de Gusmao C.X. Lu Y. Almalioglu S. Rosa C. Chen J. Wahlstrom W. Wang A. Markham and N. Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
        In <em>IEEE Robotics and Automation Letters (RAL) + IEEE ICRA</em>.
        2020.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication13786-abstract.html">
			Details <span class="sr-only" > about <em> DeepTIO: A Deep Thermal&#x2212;Inertial Odometry with Visual Hallucination  </em></span> </a> |
			<a href="/publications/publication13786.bib">
			BibTeX  <span class="sr-only" > data for <em> DeepTIO: A Deep Thermal&#x2212;Inertial Odometry with Visual Hallucination  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/11551/RAL2020_DeepTIO.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> DeepTIO: A Deep Thermal&#x2212;Inertial Odometry with Visual Hallucination  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[2] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            Distilling Knowledge From a Deep Pose Regressor Network </strong> </p>
	    <p class="publicationAuthors"> <em> Muhamad Risqi U. Saputra Pedro P. B. de Gusmao Yasin Almalioglu Andrew Markham and Niki Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
        In <em>IEEE/CVF International Conference on Computer Vision (ICCV)</em>.
        2019.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication12673-abstract.html">
			Details <span class="sr-only" > about <em> Distilling Knowledge From a Deep Pose Regressor Network  </em></span> </a> |
			<a href="/publications/publication12673.bib">
			BibTeX  <span class="sr-only" > data for <em> Distilling Knowledge From a Deep Pose Regressor Network  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/11903/PublicationFile">Download
					    

					    <span class="sr-only" > of <em> Distilling Knowledge From a Deep Pose Regressor Network  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[3] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            Learning Monocular Visual Odometry through Geometry&#x2212;Aware Curriculum Learning </strong> </p>
	    <p class="publicationAuthors"> <em> Muhamad Risqi U. Saputra Pedro P. B. de Gusmao Sen Wang Andrew Markham and Niki Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
        In <em>IEEE International Conference on Robotics and Automation (ICRA)</em>.
        2019.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication12492-abstract.html">
			Details <span class="sr-only" > about <em> Learning Monocular Visual Odometry through Geometry&#x2212;Aware Curriculum Learning  </em></span> </a> |
			<a href="/publications/publication12492.bib">
			BibTeX  <span class="sr-only" > data for <em> Learning Monocular Visual Odometry through Geometry&#x2212;Aware Curriculum Learning  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10835/_ICRA__Learning_Monocular_Visual_Odometry_through_Geometry_Aware_Curriculum_Learning.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> Learning Monocular Visual Odometry through Geometry&#x2212;Aware Curriculum Learning  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[4] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            GANVO: Unsupervised Deep Monocular Visual Odometry and Depth Estimation with Generative Adversarial Networks </strong> </p>
	    <p class="publicationAuthors"> <em> Y. Almalioglu M. R. U. Saputra P. P. de Gusmao A. Markham and N. Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
        In <em>IEEE International Conference on Robotics and Automation (ICRA)</em>.
        2019.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication12491-abstract.html">
			Details <span class="sr-only" > about <em> GANVO: Unsupervised Deep Monocular Visual Odometry and Depth Estimation with Generative Adversarial Networks  </em></span> </a> |
			<a href="/publications/publication12491.bib">
			BibTeX  <span class="sr-only" > data for <em> GANVO: Unsupervised Deep Monocular Visual Odometry and Depth Estimation with Generative Adversarial Networks  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10834/GANVO.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> GANVO: Unsupervised Deep Monocular Visual Odometry and Depth Estimation with Generative Adversarial Networks  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[5] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            DeepVO: Towards End&#x2212;to&#x2212;End Visual Odometry with Deep Recurrent Convolutional Neural Networks </strong> </p>
	    <p class="publicationAuthors"> <em> S. Wang&#x201a; R. Clark&#x201a; H. Wen and N. Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
        In <em>International Conference on Robotics and Automation</em>.
        2017.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication10904-abstract.html">
			Details <span class="sr-only" > about <em> DeepVO: Towards End&#x2212;to&#x2212;End Visual Odometry with Deep Recurrent Convolutional Neural Networks  </em></span> </a> |
			<a href="/publications/publication10904.bib">
			BibTeX  <span class="sr-only" > data for <em> DeepVO: Towards End&#x2212;to&#x2212;End Visual Odometry with Deep Recurrent Convolutional Neural Networks  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/9026/DeepVO.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> DeepVO: Towards End&#x2212;to&#x2212;End Visual Odometry with Deep Recurrent Convolutional Neural Networks  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li></ul><h2>
      Conference papers
      </h2><ul class="list-unbulleted list-unbulleted-spaced"><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[1] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            Autonomous Learning for Face Recognition in the Wild via Ambient Wireless Cues </strong> </p>
	    <p class="publicationAuthors"> <em> <u>Chris Xiaoxuan Lu</u>&#x201a; Xuan Kan&#x201a; Bowen Du&#x201a; Changhao Chen&#x201a; Hongkai Wen&#x201a; Andrew Markham&#x201a; Niki Trigoni and John Stankovic  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            The Web Conference (WWW)</em>.
        2019.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication12279-abstract.html">
			Details <span class="sr-only" > about <em> Autonomous Learning for Face Recognition in the Wild via Ambient Wireless Cues  </em></span> </a> |
			<a href="/publications/publication12279.bib">
			BibTeX  <span class="sr-only" > data for <em> Autonomous Learning for Face Recognition in the Wild via Ambient Wireless Cues  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10935/%5BWWW2019%5Dautotune.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> Autonomous Learning for Face Recognition in the Wild via Ambient Wireless Cues  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[2] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            Demo Abstract: Automatic Face Recognition Adaptation via Ambient Wireless Identifiers </strong> </p>
	    <p class="publicationAuthors"> <em> Chris Xiaoxuan Lu&#x201a; Peijun Zhao&#x201a; Bowen Du&#x201a; Hongkai Wen&#x201a; Andrew Markham&#x201a; Stefano Rosa and Niki Trignoni  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            Proceedings of the 15th ACM Conference on Embedded Network Sensor Systems (SenSys)</em>.
        2018.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication12154-abstract.html">
			Details <span class="sr-only" > about <em> Demo Abstract: Automatic Face Recognition Adaptation via Ambient Wireless Identifiers  </em></span> </a> |
			<a href="/publications/publication12154.bib">
			BibTeX  <span class="sr-only" > data for <em> Demo Abstract: Automatic Face Recognition Adaptation via Ambient Wireless Identifiers  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10589/Demo%20Abstract%20Automatic%20Face%20Recognition%20Adaptation%20via%20Ambient%20Wireless%20Identifiers.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> Demo Abstract: Automatic Face Recognition Adaptation via Ambient Wireless Identifiers  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[3] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            IONet: Learning to Cure the Curse of Drift in Inertial Odometry </strong> </p>
	    <p class="publicationAuthors"> <em> Changhao Chen&#x201a; Chris Xiaoxuan Lu&#x201a; Andrew Markham and Niki Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            The Thirty&#x2212;Second AAAI Conference on Artificial Intelligence (AAAI&#x2212;18)</em>.
        2018.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication11583-abstract.html">
			Details <span class="sr-only" > about <em> IONet: Learning to Cure the Curse of Drift in Inertial Odometry  </em></span> </a> |
			<a href="/publications/publication11583.bib">
			BibTeX  <span class="sr-only" > data for <em> IONet: Learning to Cure the Curse of Drift in Inertial Odometry  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/9853/Learning%20to%20Cure%20the%20Curse%20of%20Drift%20in%20Inertial%20Odometry.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> IONet: Learning to Cure the Curse of Drift in Inertial Odometry  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[4] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            Snoopy: Sniffing Your Smartwatches Passwords via Deep Sequence Learning </strong> </p>
	    <p class="publicationAuthors"> <em> Chris Xiaoxuan Lu&#x201a; Bowen Du&#x201a; Hongkai Wen&#x201a; Sen Wang&#x201a; Andrew Markham&#x201a; Ivan Martinovic&#x201a; Yiran Shen and Niki Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp)</em>.
        2018.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication11247-abstract.html">
			Details <span class="sr-only" > about <em> Snoopy: Sniffing Your Smartwatches Passwords via Deep Sequence Learning  </em></span> </a> |
			<a href="/publications/publication11247.bib">
			BibTeX  <span class="sr-only" > data for <em> Snoopy: Sniffing Your Smartwatches Passwords via Deep Sequence Learning  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10773/slides.key">Download
					    slides.key

					    <span class="sr-only" > of <em> Snoopy: Sniffing Your Smartwatches Passwords via Deep Sequence Learning  </em></span>

					    </a>


				    

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10772/%5BUbiComp2018%5Dsnoopy.pdf">Download
					    [UbiComp2018]snoopy.pdf

					    <span class="sr-only" > of <em> Snoopy: Sniffing Your Smartwatches Passwords via Deep Sequence Learning  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[5] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            VINet: Visual Inertial Odometry as a Sequence to Sequence Learning Problem </strong> </p>
	    <p class="publicationAuthors"> <em> R. Clark&#x201a; S. Wang&#x201a; H. Wen&#x201a; A. Markham and N. Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            AAAI Conference on Artificial Intelligence (AAAI)</em>.
        2017.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication10903-abstract.html">
			Details <span class="sr-only" > about <em> VINet: Visual Inertial Odometry as a Sequence to Sequence Learning Problem  </em></span> </a> |
			<a href="/publications/publication10903.bib">
			BibTeX  <span class="sr-only" > data for <em> VINet: Visual Inertial Odometry as a Sequence to Sequence Learning Problem  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/9032/VINet.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> VINet: Visual Inertial Odometry as a Sequence to Sequence Learning Problem  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[6] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            SCAN: Learning Speaker Identity From Noisy Sets of Sensor Data </strong> </p>
	    <p class="publicationAuthors"> <em> <u>Chris Xiaoxuan Lu</u>&#x201a; Hongkai Wen&#x201a; Sen Wang&#x201a; Andrew Markham and Niki Trigoni  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN)</em>.
        2017.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication10934-abstract.html">
			Details <span class="sr-only" > about <em> SCAN: Learning Speaker Identity From Noisy Sets of Sensor Data  </em></span> </a> |
			<a href="/publications/publication10934.bib">
			BibTeX  <span class="sr-only" > data for <em> SCAN: Learning Speaker Identity From Noisy Sets of Sensor Data  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/10079/scan.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> SCAN: Learning Speaker Identity From Noisy Sets of Sensor Data  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[7] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            VidLoc: A Deep Spatio&#x2212;Temporal Model for 6&#x2212;DoF Video&#x2212;Clip Relocalization </strong> </p>
	    <p class="publicationAuthors"> <em> Ronald Clark&#x201a; Sen Wang&#x201a; Andrew Markham&#x201a; Niki Trigoni and Hongkai Wen  </em> </p>
	
	<p class="publicationDetails"> 
            In  
            <em>
            Computer Vision and Pattern Recognition (CVPR)</em>.
        2017.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication11686-abstract.html">
			Details <span class="sr-only" > about <em> VidLoc: A Deep Spatio&#x2212;Temporal Model for 6&#x2212;DoF Video&#x2212;Clip Relocalization  </em></span> </a> |
			<a href="/publications/publication11686.bib">
			BibTeX  <span class="sr-only" > data for <em> VidLoc: A Deep Spatio&#x2212;Temporal Model for 6&#x2212;DoF Video&#x2212;Clip Relocalization  </em></span>   </a>
			
			

				     |
                			<a 
                			    href="http://www.cs.ox.ac.uk/files/9725/VidLoc.pdf">Download
					     (pdf)

					    <span class="sr-only" > of <em> VidLoc: A Deep Spatio&#x2212;Temporal Model for 6&#x2212;DoF Video&#x2212;Clip Relocalization  </em></span>

					    </a>


				    
		
		</div> 
	</div> 
	
        </li><li>
	
	<div class="row">
		
			<div class="col-xs-1 text-right">
				[8] 
			</div>

		
        
		<div
			class="col-xs-11"
		>
			
	
	<p class="publicationTitle"> <strong>
            "Gabor filter based image representation for object classification." </strong> </p>
	    <p class="publicationAuthors"> <em> Gianluca. Rizvi S.T.H.; Cabodi Gianpiero; P. B. de Gusmao Pedro; Francini  </em> </p>
	
	<p class="publicationDetails"> 2016.
        </p>
	
        		<p class="publicationlinks"> <a
        
            		href="/publications/publication11690-abstract.html">
			Details <span class="sr-only" > about <em> "Gabor filter based image representation for object classification."  </em></span> </a> |
			<a href="/publications/publication11690.bib">
			BibTeX  <span class="sr-only" > data for <em> "Gabor filter based image representation for object classification."  </em></span>   </a>
			
			 | <a
		href="http://dx.doi.org/10.1109/CoDIT.2016.7593635">
		DOI (10.1109/CoDIT.2016.7593635)
		</a> 
		
		</div> 
	</div> 
	
        </li></ul></div><div class="col-xs-12 col-md-3 col-md-pull-9 side-menu-col">[an error occurred while processing this directive]</div></div></div>[an error occurred while processing this directive]