BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Europe/Berlin BEGIN:DAYLIGHT DTSTART:20260329T030000 TZOFFSETFROM:+0100 TZOFFSETTO:+0200 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3 TZNAME:CEST END:DAYLIGHT BEGIN:STANDARD DTSTART:20261025T020000 TZOFFSETFROM:+0200 TZOFFSETTO:+0100 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:CET END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260509T153915Z UID:8E0F3935-4B3D-4F00-8500-D940314E5773 DTSTART;TZID=Europe/Berlin:20260505T171500 DTEND;TZID=Europe/Berlin:20260505T181500 DESCRIPTION:We would like to invite all German IEEE members to the upcoming LMAG Germany Technical Meeting with a technical presentation on AI\n\n05 May 2026\, 17:15 - 18:15 CEST\n\nWebEX link: https://uni-siegen.webex.com/ uni-siegen/j.php?MTID=mb3d90bf60d6e2801f9af973fea48f890\n\n(Same WebEX ses sion link as used for preceding LMAG Germany organizational meeting)\n\nAg enda proposal:\n\n- Call to order\n- Minutes of the last meeting\n\n- Pres entation of Prof. Michael Möller\, Uni Siegen:\n"Learning to Sense - Join t Optimization of Imaging System and Machine Learning Parameters"\n- Quest ions\, answers\, and discussions\n- Next meeting on July 7th\n- AOB\n\nAbo ut the presentation of Prof. Michael Möller:\n\nLearning to Sense - Joint Optimization of Imaging System and Machine Learning Parameters\n\nAbstrac t: Traditional imaging systems are designed to maximize the visual quality of the recorded images for human inspection on a screen. Yet nowadays\, a n increasing number of imaging systems provide data for automatic analysis with machine learning approaches only. This gives rise to the idea of opt imizing sensor system parameters jointly with neural network parameters in such a way that the recorded data is optimal for subsequent machine learn ing-based analysis. I will illustrate this idea in several applications\, including optimizing for non-uniform pixel layouts on a sensor and and hig hly efficient video data recording via learned quantization.\n\nAgenda: \n Agenda proposal:\n\n- Call to order\n- Minutes of the last meeting\n- Pres entation of Prof. Michael Möller\, Uni Siegen:\n"Learning to Sense - Join t Optimization of Imaging System and Machine Learning Parameters"\n- Quest ions\, answers\, and discussions\n- Next meeting on July 7th\n- AOB\n\nVir tual: https://events.vtools.ieee.org/m/546218 LOCATION:Virtual: https://events.vtools.ieee.org/m/546218 ORGANIZER:max.riegel@ieee.org SEQUENCE:18 SUMMARY:LMAG Germany Technical Meeting URL;VALUE=URI:https://events.vtools.ieee.org/m/546218 X-ALT-DESC:Description: <br /><p>We would like to invite all German IEEE me mbers to the upcoming LMAG Germany Technical Meeting with a technical pres entation on AI</p>\n<p>05 May 2026\, 17:15 - 18:15 CEST</p>\n<p>WebEX link : <a href="https://uni-siegen.webex.com/uni-siegen/j.php?MTID=mb3d90bf60d6 e2801f9af973fea48f890">https://uni-siegen.webex.com/uni-siegen/j.php?MTID= mb3d90bf60d6e2801f9af973fea48f890</a></p>\n<p>(Same WebEX session link as used for preceding LMAG Germany organizational meeting)</p>\n<p>&nbsp\;</p >\n<p>Agenda proposal:</p>\n<ul>\n<li>Call to order</li>\n<li>Minutes of t he last meeting<span style="font-family: 'Calibri'\,sans-serif\; color: bl ack\;"><br></span></li>\n<li>Presentation of Prof. Michael M&ouml\;ller\, Uni Siegen:<br>"Learning to Sense - Joint Optimization of Imaging System a nd Machine Learning Parameters"</li>\n<li>Questions\, answers\, and discus sions</li>\n<li>Next meeting on July 7th</li>\n<li>AOB</li>\n</ul>\n<p>&nb sp\;</p>\n<p>About the presentation of Prof. Michael M&ouml\;ller:</p>\n<p ><strong>Learning to Sense - Joint Optimization of Imaging System and Mach ine Learning Parameters</strong></p>\n<p>Abstract: Traditional imaging sys tems are designed to maximize the visual quality of the recorded images fo r human inspection on a screen. Yet nowadays\, an increasing number of ima ging systems provide data for automatic analysis with machine learning app roaches only. This gives rise to the idea of optimizing sensor system para meters jointly with neural network parameters in such a way that the recor ded data is optimal for subsequent machine learning-based analysis. I will illustrate this idea in several applications\, including optimizing for n on-uniform pixel layouts on a sensor and and highly efficient video data r ecording via learned quantization.</p>\n<p>&nbsp\;</p><br /><br />Agenda: <br /><p>Agenda proposal:</p>\n<ul>\n<li>Call to order</li>\n<li>Minutes o f the last meeting</li>\n<li>Presentation of Prof. Michael M&ouml\;ller\, Uni Siegen:<br>"Learning to Sense - Joint Optimization of Imaging System a nd Machine Learning Parameters"</li>\n<li>Questions\, answers\, and discus sions</li>\n<li>Next meeting on July 7th</li>\n<li>AOB</li>\n</ul> END:VEVENT END:VCALENDAR