{"id":22249,"date":"2025-03-03T13:32:05","date_gmt":"2025-03-03T12:32:05","guid":{"rendered":"https:\/\/www.lites.tf.fau.de\/?p=22249"},"modified":"2025-05-05T16:51:39","modified_gmt":"2025-05-05T14:51:39","slug":"erstellung-eines-synthetischen-datensatzes-fuer-ein-cnn-basiertes-parkdeck-training","status":"publish","type":"post","link":"https:\/\/www.lites.tf.fau.de\/en\/2025\/03\/03\/erstellung-eines-synthetischen-datensatzes-fuer-ein-cnn-basiertes-parkdeck-training\/","title":{"rendered":"Creation of a Synthetic Dataset in Blender for CNN-Based Parking Deck Training"},"content":{"rendered":"

Master\u2019s Thesis \/ Bachelor\u2019s Thesis \/ Research Internship<\/h3>\n

Description<\/strong><\/h3>\n

The use of synthetic datasets<\/strong> in computer vision<\/strong> is becoming increasingly important, especially for training neural networks<\/strong>. This project aims to create a digital 3D model of a parking deck<\/strong> in Blender to realistically simulate various weather conditions and environmental factors<\/strong>. The generated dataset will serve as the basis for training a Convolutional Neural Network (CNN)<\/strong>, which will subsequently be tested and evaluated on real images of the parking deck.<\/p>\n

Research Questions<\/strong><\/h3>\n
    \n
  • How realistically can synthetic datasets be generated using Blender?<\/li>\n
  • To what extent can neural networks trained on synthetic data be transferred to real-world scenarios?<\/li>\n
  • Which environmental factors have the greatest impact on object recognition in parking decks?<\/li>\n<\/ul>\n

    Objectives of the Work<\/strong><\/h3>\n

    The objective of this project is to create a synthetic dataset<\/strong> that represents various environmental conditions (e.g., rain, fog, shadows)<\/strong> and improves the generalization capability of a CNN<\/strong>. Furthermore, the study will analyze how well the trained network can be applied to real-world images<\/strong>.<\/p>\n

    Work Packages<\/strong><\/h3>\n
      \n
    • Modeling a 3D parking deck<\/strong> in Blender with realistic structures<\/li>\n
    • Simulating different environmental conditions<\/strong> (lighting variations, weather effects)<\/li>\n
    • Rendering images from multiple perspectives<\/strong> to generate the dataset<\/li>\n
    • Training a CNN<\/strong> using the synthetic dataset<\/li>\n
    • Testing and evaluating<\/strong> the network on real parking deck images<\/li>\n<\/ul>\n

      Your Skills<\/strong><\/h3>\n
        \n
      • Experience in 3D modeling<\/strong> (e.g., Blender)<\/li>\n
      • Basic knowledge of Deep Learning and Computer Vision<\/strong><\/li>\n
      • Programming skills in Python<\/strong> (e.g., TensorFlow, PyTorch)<\/li>\n
      • Interest in synthetic datasets and image processing<\/strong><\/li>\n<\/ul>\n
        \n

        Es konnte kein Kontakteintrag mit der angegebenen ID 3923 gefunden werden.<\/div>
        \n
        <\/div><\/p>\n","protected":false},"excerpt":{"rendered":"

        Master\u2019s Thesis \/ Bachelor\u2019s Thesis \/ Research Internship Description The use of synthetic datasets in computer vision is becoming increasingly important, especially for training neural networks. This project aims to create a digital 3D model of a parking deck in Blender to realistically simulate various weather conditions and environmental factors. The generated dataset will serve […]<\/p>\n","protected":false},"author":4019,"featured_media":22033,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_rrze_cache":"enabled","_rrze_multilang_single_locale":"en_US","_rrze_multilang_single_source":"https:\/\/www.lites.tf.fau.de\/?p=22245","footnotes":""},"categories":[581,226],"tags":[],"workflow_usergroup":[],"class_list":["post-22249","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-fp_eng","category-offene_stud_arbeiten_eng","en-US"],"_links":{"self":[{"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/posts\/22249","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/users\/4019"}],"replies":[{"embeddable":true,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/comments?post=22249"}],"version-history":[{"count":4,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/posts\/22249\/revisions"}],"predecessor-version":[{"id":22626,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/posts\/22249\/revisions\/22626"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/media\/22033"}],"wp:attachment":[{"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/media?parent=22249"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/categories?post=22249"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/tags?post=22249"},{"taxonomy":"workflow_usergroup","embeddable":true,"href":"https:\/\/www.lites.tf.fau.de\/wp-json\/wp\/v2\/workflow_usergroup?post=22249"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}