BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20240310T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20241103T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20240926T193749Z UID:8AF25B57-6057-4275-A9DE-EA53B221A34E DTSTART;TZID=America/New_York:20240926T140000 DTEND;TZID=America/New_York:20240926T150000 DESCRIPTION:PreFab: Image-to-Image Computer Vision for Advanced Nanophotoni c Fabrication and Design\n\nAbstract: PreFab is an image-to-image computer vision system designed to predict and correct nanofabrication variations in integrated photonic circuits. Utilizing convolutional neural networks ( CNNs) trained on GDS design files and scanning electron microscope (SEM) i mages\, PreFab enables precise transformations such as fabrication variati on prediction\, automatic design correction\, SEM segmentation\, and SEM s tyling. This system enhances the manufacturability and fidelity of complex photonic circuits by anticipating deviations like erosion\, dilation\, an d edge uncertainties\, and compensating for these during the design phase. Additionally\, it facilitates post-fabrication analysis through SEM segme ntation\, allowing for accurate performance analysis. PreFab's methodology aims to minimize fabrication cycles\, enhance design fidelity\, and adapt to diverse nanophotonic manufacturing techniques\, positioning it as a us eful tool for the advancement of next-generation photonic technologies.\n\ n------------------------------------------------------------------------\ n\nPreFab : Vision par ordinateur image à image pour la fabrication et la conception nanophotoniques avancées\n\nRésumé : PreFab est un système de vision par ordinateur image à image conçu pour prédire et corriger les variations de nanofabrication dans les circuits photoniques intégrés . En utilisant des réseaux neuronaux convolutionnels (CNN) formés sur de s fichiers de conception GDS et des images de microscope électronique à balayage (SEM)\, PreFab permet des transformations précises telles que la prédiction des variations de fabrication\, la correction automatique de la conception\, la segmentation SEM et le style SEM. Ce système améliore la fabricabilité et la fidélité des circuits photoniques complexes en anticipant les écarts tels que l'érosion\, la dilatation et les incertit udes de bord\, et en les compensant pendant la phase de conception. De plu s\, il facilite l'analyse post-fabrication grâce à la segmentation SEM\, permettant une analyse précise des performances. La méthodologie de Pre Fab vise à minimiser les cycles de fabrication\, à améliorer la fidéli té de la conception et à s'adapter à diverses techniques de fabrication nanophotonique\, le positionnant comme un outil utile pour l'avancement d es technologies photoniques de nouvelle génération.\n\n[]\n\nDr. Dusan G ostimirovic (PreFab Photonics)\n\nAbout / A propos\n\nThe High Throughput and Secure Networks (HTSN) Challenge program is hosting regular virtual se minar series to promote scientific information sharing\, discussions\, and interactions between researchers.\n\nhttps://nrc.canada.ca/en/research-de velopment/research-collaboration/programs/high-throughput-secure-networks- challenge-program\n\nLe programme Réseaux Sécurisés à Haut Débit (RSH D) organise régulièrement des séries de séminaires virtuels pour promo uvoir le partage d’informations scientifiques\, les discussions et les i nteractions entre chercheurs.\n\nhttps://nrc.canada.ca/fr/recherche-develo ppement/recherche-collaboration/programmes/programme-defi-reseaux-securise s-haut-debit\n\nCo-sponsored by: National Research Council\, Canada. Opton ique.\n\nSpeaker(s): Dusan Gostimirovic\, \n\nVirtual: https://events.vtoo ls.ieee.org/m/433675 LOCATION:Virtual: https://events.vtools.ieee.org/m/433675 ORGANIZER:Alex.Poungoue@ieee.org SEQUENCE:29 SUMMARY:PreFab: Image-to-Image Computer Vision for Advanced Nanophotonic Fa brication and Design URL;VALUE=URI:https://events.vtools.ieee.org/m/433675 X-ALT-DESC:Description: <br /><p><em><strong>PreFab: Image-to-Image Compute r Vision for Advanced Nanophotonic Fabrication and Design</strong></em></p >\n<p class="MsoNormal"><strong>Abstract</strong>: PreFab is an image-to-i mage computer vision system designed to predict and correct nanofabricatio n variations in integrated photonic circuits. Utilizing convolutional neur al networks (CNNs) trained on GDS design files and scanning electron micro scope (SEM) images\, PreFab enables precise transformations such as fabric ation variation prediction\, automatic design correction\, SEM segmentatio n\, and SEM styling. This system enhances the manufacturability and fideli ty of complex photonic circuits by anticipating deviations like erosion\, dilation\, and edge uncertainties\, and compensating for these during the design phase. Additionally\, it facilitates post-fabrication analysis thro ugh SEM segmentation\, allowing for accurate performance analysis. PreFab' s methodology aims to minimize fabrication cycles\, enhance design fidelit y\, and adapt to diverse nanophotonic manufacturing techniques\, positioni ng it as a useful tool for the advancement of next-generation photonic tec hnologies.</p>\n<p>------------------------------------------------------- -----------------</p>\n<p><em><strong>PreFab : Vision par ordinateur image &agrave\; image pour la fabrication et la conception nanophotoniques avan c&eacute\;es</strong></em></p>\n<p class="MsoNormal"><strong><span lang="F R-CA">R&eacute\;sum&eacute\; : </span></strong><span lang="FR-CA">PreFab e st un syst&egrave\;me de vision par ordinateur image &agrave\; image con&c cedil\;u pour pr&eacute\;dire et corriger les variations de nanofabricatio n dans les circuits photoniques int&eacute\;gr&eacute\;s. En utilisant des r&eacute\;seaux neuronaux convolutionnels (CNN) form&eacute\;s sur des fi chiers de conception GDS et des images de microscope &eacute\;lectronique &agrave\; balayage (SEM)\, PreFab permet des transformations pr&eacute\;ci ses telles que la pr&eacute\;diction des variations de fabrication\, la co rrection automatique de la conception\, la segmentation SEM et le style SE M. Ce syst&egrave\;me am&eacute\;liore la fabricabilit&eacute\; et la fid& eacute\;lit&eacute\; des circuits photoniques complexes en anticipant les &eacute\;carts tels que l'&eacute\;rosion\, la dilatation et les incertitu des de bord\, et en les compensant pendant la phase de conception. De plus \, il facilite l'analyse post-fabrication gr&acirc\;ce &agrave\; la segmen tation SEM\, permettant une analyse pr&eacute\;cise des performances. La m &eacute\;thodologie de PreFab vise &agrave\; minimiser les cycles de fabri cation\, &agrave\; am&eacute\;liorer la fid&eacute\;lit&eacute\; de la con ception et &agrave\; s'adapter &agrave\; diverses techniques de fabricatio n nanophotonique\, le positionnant comme un outil utile pour l'avancement des technologies photoniques de nouvelle g&eacute\;n&eacute\;ration.</span ></p>\n<p class="MsoNormal"><img src="https://events.vtools.ieee.org/vtool s_ui/media/display/8cc5ea4b-4a59-4271-aca1-d3514e5040b6" alt="" width="241 " height="308"></p>\n<p><span style="color: rgb(0\, 0\, 0)\;"><strong><spa n style="text-decoration: underline\;">Dr. Dusan Gostimirovic (PreFab Phot onics)</span></strong></span></p>\n<p><span style="text-decoration: underl ine\;">About / A propos</span></p>\n<p class="MsoNormal">The High Throughp ut and Secure Networks (HTSN) Challenge program is hosting regular virtual seminar series to promote scientific information sharing\, discussions\, and interactions between researchers.</p>\n<p class="MsoNormal"><a href="h ttps://nrc.canada.ca/en/research-development/research-collaboration/progra ms/high-throughput-secure-networks-challenge-program" target="_blank" rel= "noopener" data-saferedirecturl="https://www.google.com/url?q=https://nrc. canada.ca/en/research-development/research-collaboration/programs/high-thr oughput-secure-networks-challenge-program&amp\;source=gmail&amp\;ust=17147 77369757000&amp\;usg=AOvVaw2978XFVhIMFshBlYbS61GG">https://nrc.canada.ca/e n/<wbr>research-development/research-<wbr>collaboration/programs/high-<wbr >throughput-secure-networks-<wbr>challenge-program</a></p>\n<p class="MsoN ormal"><span lang="FR-CA">Le programme R&eacute\;seaux S&eacute\;curis&eac ute\;s &agrave\; Haut D&eacute\;bit (RSHD) organise r&eacute\;guli&egrave\ ;rement des s&eacute\;ries de s&eacute\;minaires virtuels pour promouvoir le partage d&rsquo\;informations scientifiques\, les discussions et les in teractions entre chercheurs.</span></p>\n<p class="MsoNormal"><a href="htt ps://nrc.canada.ca/fr/recherche-developpement/recherche-collaboration/prog rammes/programme-defi-reseaux-securises-haut-debit" target="_blank" rel="n oopener" data-saferedirecturl="https://www.google.com/url?q=https://nrc.ca nada.ca/fr/recherche-developpement/recherche-collaboration/programmes/prog ramme-defi-reseaux-securises-haut-debit&amp\;source=gmail&amp\;ust=1714777 369758000&amp\;usg=AOvVaw1jVTR0SH26o88wijXKtYGT"><span lang="FR-CA">https: //nrc.canada.ca/fr/<wbr>recherche-developpement/<wbr>recherche-collaborati on/<wbr>programmes/programme-defi-<wbr>reseaux-securises-haut-debit</span> </a></p> END:VEVENT END:VCALENDAR