{"id":909,"date":"2025-12-01T19:56:51","date_gmt":"2025-12-01T17:56:51","guid":{"rendered":"https:\/\/digital-forensics.mst.duth.gr\/?page_id=909"},"modified":"2025-12-01T19:58:04","modified_gmt":"2025-12-01T17:58:04","slug":"edgeai4uav","status":"publish","type":"page","link":"https:\/\/digital-forensics.mst.duth.gr\/en\/edgeai4uav\/","title":{"rendered":"edgeAI4UAV"},"content":{"rendered":"<div data-elementor-type=\"wp-page\" data-elementor-id=\"909\" class=\"elementor elementor-909\">\n\t\t\t\t\t\t<section class=\"has_eae_slider elementor-section elementor-top-section elementor-element elementor-element-2bdad0e1 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-eae-slider=\"92010\" data-id=\"2bdad0e1\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"has_eae_slider elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-43245eaa\" data-eae-slider=\"41177\" data-id=\"43245eaa\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-5dd7e171 elementor-widget elementor-widget-text-editor\" data-id=\"5dd7e171\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<div style=\"padding: 3rem;\"><table align=\"\"><tbody><tr><td style=\"text-align: center; vertical-align: middle;\">\u00a0<\/td><td style=\"text-align: center; vertical-align: middle;\">\u00a0<\/td><\/tr><\/tbody><\/table><p><a href=\"https:\/\/chattostudy.tu-sofia.bg\/\"><strong><span lang=\"BG\">Computer Vision and AI Algorithms Edge Computation on UAVs (edgeAI4UAV)<\/span><\/strong><\/a><\/p><ul><li><p data-start=\"203\" data-end=\"719\">The project <em data-start=\"215\" data-end=\"290\">\u201cComputer Vision and AI Algorithms Edge Computation on UAVs (edgeAI4UAV)\u201d<\/em> has been selected as one of the five research initiatives funded through the <strong data-start=\"368\" data-end=\"444\">AI4Media \u2013 A European Excellence Centre for Media, Society and Democracy<\/strong>. It is implemented by members of the laboratory under the scientific supervision of <strong data-start=\"529\" data-end=\"563\">Professor Dr. Vasileios Chatzi<\/strong>, in collaboration with researchers from the Departments of Management Science &amp; Technology and Computer Science of the <strong data-start=\"683\" data-end=\"718\">Democritus University of Thrace<\/strong>.<\/p><h3 data-start=\"721\" data-end=\"737\"><strong data-start=\"725\" data-end=\"737\">Overview<\/strong><\/h3><p data-start=\"738\" data-end=\"1145\">The primary goal of edgeAI4UAV is to develop innovative <strong data-start=\"794\" data-end=\"813\">computer vision<\/strong> and <strong data-start=\"818\" data-end=\"861\">Artificial Intelligence (deep learning)<\/strong> algorithms capable of running directly on-board Unmanned Aerial Vehicles (UAVs) through <strong data-start=\"950\" data-end=\"968\">edge computing<\/strong> techniques. The system focuses on real-time detection and tracking of people and moving objects, enabling autonomous cinematography tasks and event-driven visual documentation.<\/p><h3 data-start=\"1147\" data-end=\"1173\"><strong data-start=\"1151\" data-end=\"1173\">Technical Approach<\/strong><\/h3><p data-start=\"1174\" data-end=\"1197\">The project integrates:<\/p><ul data-start=\"1199\" data-end=\"1472\"><li data-start=\"1199\" data-end=\"1302\"><p data-start=\"1201\" data-end=\"1302\">A <strong data-start=\"1203\" data-end=\"1227\">stereo camera system<\/strong>, providing depth information via lightweight stereo-to-depth algorithms.<\/p><\/li><li data-start=\"1303\" data-end=\"1378\"><p data-start=\"1305\" data-end=\"1378\">An <strong data-start=\"1308\" data-end=\"1341\">embedded edge processing unit<\/strong> for real-time onboard computation.<\/p><\/li><li data-start=\"1379\" data-end=\"1472\"><p data-start=\"1381\" data-end=\"1472\">A <strong data-start=\"1383\" data-end=\"1398\">WiFi module<\/strong> allowing the UAV to transmit selected images to a server while in flight.<\/p><\/li><\/ul><p data-start=\"1474\" data-end=\"1714\">The computer vision algorithms are designed to detect and track moving targets, while their outputs are utilized by an <strong data-start=\"1593\" data-end=\"1628\">embedded decision-making module<\/strong>. This module dynamically adjusts UAV navigation in real time, enabling tasks such as:<\/p><ul data-start=\"1716\" data-end=\"1917\"><li data-start=\"1716\" data-end=\"1787\"><p data-start=\"1718\" data-end=\"1787\">Following a specific moving person or object (e.g., actor, animal).<\/p><\/li><li data-start=\"1788\" data-end=\"1838\"><p data-start=\"1790\" data-end=\"1838\">Changing viewpoint angles (side, front, etc.).<\/p><\/li><li data-start=\"1839\" data-end=\"1917\"><p data-start=\"1841\" data-end=\"1917\">Approaching or distancing from the target depending on mission requirements.<\/p><\/li><\/ul><p data-start=\"1919\" data-end=\"2081\">Through this integrated perception-and-control framework, the UAV can autonomously execute complex cinematographic missions without continuous human intervention.<\/p><h3 data-start=\"2083\" data-end=\"2097\"><strong data-start=\"2087\" data-end=\"2097\">Impact<\/strong><\/h3><p data-start=\"2098\" data-end=\"2431\">The project advances the field of autonomous aerial cinematography by combining deep learning\u2013based perception with classical path-planning and control methods. It enables high-quality visual content acquisition in dynamic environments, while demonstrating the potential of <strong data-start=\"2372\" data-end=\"2383\">edge AI<\/strong> for real-time decision-making in UAV platforms.<\/p><h3 data-start=\"2433\" data-end=\"2458\"><strong data-start=\"2437\" data-end=\"2458\">Main Deliverables<\/strong><\/h3><ul data-start=\"2459\" data-end=\"2730\"><li data-start=\"2459\" data-end=\"2564\"><p data-start=\"2461\" data-end=\"2564\">Real-time computer vision algorithms for <strong data-start=\"2502\" data-end=\"2534\">human detection and tracking<\/strong> from an onboard UAV camera.<\/p><\/li><li data-start=\"2565\" data-end=\"2646\"><p data-start=\"2567\" data-end=\"2646\">A <strong data-start=\"2569\" data-end=\"2608\">decision-support and control system<\/strong> enabling autonomous UAV navigation.<\/p><\/li><li data-start=\"2647\" data-end=\"2730\"><p data-start=\"2649\" data-end=\"2730\">A fully functional <strong data-start=\"2668\" data-end=\"2685\">UAV prototype<\/strong> with integrated human-tracking capabilities.<\/p><\/li><\/ul><h3 data-start=\"2732\" data-end=\"2747\"><strong data-start=\"2736\" data-end=\"2747\">Funding<\/strong><\/h3><p data-start=\"2748\" data-end=\"2975\">The edgeAI4UAV project has indirectly received funding from the European Union\u2019s <strong data-start=\"2829\" data-end=\"2845\">Horizon 2020<\/strong> Research and Innovation Programme through the <strong data-start=\"2892\" data-end=\"2917\">AI4Media Open Call #1<\/strong>, under the AI4Media project (Grant Agreement No. 951911).<\/p><\/li><\/ul><hr \/><p style=\"margin: 0cm; line-height: 119%;\">\u00a0<\/p><\/div>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>","protected":false},"excerpt":{"rendered":"<p>\u00a0 \u00a0 Computer Vision and AI Algorithms Edge Computation on UAVs (edgeAI4UAV) The project \u201cComputer Vision and AI Algorithms Edge Computation on UAVs (edgeAI4UAV)\u201d has been selected as one of the five research initiatives funded through the AI4Media \u2013 A European Excellence Centre for Media, Society and Democracy. It is implemented by members of the&hellip; <br \/> <a class=\"read-more\" href=\"https:\/\/digital-forensics.mst.duth.gr\/en\/edgeai4uav\/\">Read more<\/a><\/p>","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"page-templates\/page-with-right-sidebar.php","meta":{"footnotes":""},"class_list":["post-909","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/pages\/909","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/comments?post=909"}],"version-history":[{"count":5,"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/pages\/909\/revisions"}],"predecessor-version":[{"id":920,"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/pages\/909\/revisions\/920"}],"wp:attachment":[{"href":"https:\/\/digital-forensics.mst.duth.gr\/en\/wp-json\/wp\/v2\/media?parent=909"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}