Welcome to the Chair for Machine Learning?and Computer Vision.

?

The Chair of Machine Learning and Computer Vision investigates methods for machine learning and vision, as well as perception in general. In addition to researching and developing methods specifically tailored to image, video, sensor, and multimedia data (data from various sources and types), the chair also explores data mining and analytics methods for production data, as sensor data from these production environments also represents a form of perception. A significant focus is the recognition of people, their poses and actions in 2D and 3D, the prediction of their future movements, and the evaluation of movement quality (e.g., in long jump/triple jump, swimming, and ski jumping for our Olympic athletes).

The chair also develops automated methods for recognizing medical findings and for automatically generating textual reports (report generation) from medical data (MRI scans, colonoscopy videos). A third focus of the chair is the research of methods for generating new training examples through synthesis using Generative Adversarial Networks (GANs). These methods allow the synthesis of the large number of training examples required for deep neural networks from a small domain-specific training set or additionally from a large, extra-domain training set, thus going far beyond the possibilities of simple data augmentation.

• Applied Machine Learning (e.g., deep neural networks, Bayesian networks, random forests)
• Computer Vision, including multimodal detection, localization, and recognition of people, objects, and events, as well as industrial visual and multimodal material inspection
• Data mining of multimodal data (e.g., production data)
• Indexing and searching in images, videos, and multimodal documents
• Generation of new training examples through synthesis using Generative Adversarial Networks (GANs) from a small domain-specific training set or additionally from a large, non-domain-specific training set

Bachelor?? s degree: Foundations of Multimedia, Foundations of Signal Processing and Machine Learning, Multimedia Project

Master? s degree: Machine learning and Computer Vision, Advanced?Machine learning and Computer Vision, Foundation Models in Deep Learning, Advanced Deep Learning

• An above-average academic degree (master’s or comparable) in Computer Science, Electrical Engineering, Mathematics, Physics, Data Science, or related fields,
• Passion for one of the research areas of the chair (e.g., Computer Vision, 3D Vision, Machine Learning, Physics-informed Networks, 3D Human Pose and Mesh Estimation, Spatio-Temporal Prediction) and enthusiasm and perseverance to research in this area,
• Ideally, experience in the areas of computer vision and machine learning from your studies,
• Good programming skills in a common high-level language (e.g., Python),
• Ideally, practical experience with tools for machine learning such as PyTorch, JAX, TensorFlow, or Sci-Kit Learn,
• A high degree of self-organization and initiative,
• Good German and English language skills,
• Active support of our teaching offer,
• Good communication and teamwork skills."