Research in Computer Vision

Basic Research in Computer Vision

Unsupervised Video Understanding by Reconciliation of Posture Similarities

Understanding human activity and being able to explain it in detail surpasses mere action classification by far in both complexity and value. The challenge is thus to describe an activity on the basis of its most fundamental constituents, the individual postures and their distinctive transitions. Read more

Randomized Max-Margin Compositions

A main theme in object detection are currently discriminative part-based models. The powerful model that combines all parts is then typically only feasible for few constituents, which are in turn iteratively trained to make them as strong as possible. Read more

Voting by Grouping Dependent Parts

The complexity of multi-scale, category level object detection in cluttered scenes is handled efficiently by Hough voting methods. The primary weakness of this approach is however that mutually dependent local observations are independently voting for intrinsically global object properties such as object scale. Read more.

Beyond Straight Lines - Object Detection Using Curvature

Visual object detection in cluttered scenes is one of the key problems of computer vision. Localizing all instances of an object category is highly challenging due to the large intra-class variability. Finding a common model for all the widely diverse class instances thus poses a major difficulty. Read more

Video Parsing for Abnormality Detection

Object detection is a difficult problem in computer vision because of very large intraclass variability, object articulation and view-point dependence. However, detecting abnormal objects in videos is even harder because of an enormous number of ways for an object to appear or behave abnormally. Read more

Contour-based Object Detection

Contour-based representations have a long history in object recognition and computer vision. Considerable effort was spent in the past matching geometric shape models of objects to image contours [4, 5, 9]. Read more

Max-Margin Regularization for Chamfer Matching

Chamfer matching is an effective and widely used technique for detecting objects or parts thereof by their shape. However, a serious limitation is its susceptibility to background clutter. Read more

Visual Recognition using Embedded Feature Selection for Curvature Self-Similarity

Category-Level object detection has a crucial need for informative object representations. This demand has led to feature descriptors of ever increasing dimensionality like co-occurrence statistics and self-similarity. Read more

From Meaningful Contours to Discriminative Object Shape

Shape is a natural, highly prominent characteristic of objects that human vision utilizes everyday. But despite its expressiveness, shape poses significant challenges for category-level object detection in cluttered scenes. Read more

Computer Vision in the Humanities

Computer-Assisted Detection and Analysis of Medieval Legal Gestures

In this project we seek to better understand the purpose and origins of legal gestures depicted in medieval manuscripts. Our approach is based on four illustrated manuscripts of Eike von Repgow's Sachsenspiegel (Mirror of the Saxons), one of the oldest manuscripts on German law. Read more

Drawing Process Reconstruction in Medieval Images

In the course of the reception of the Middle Ages, many outstanding medieval manuscripts were copied by hand hundreds of years after their original production. Copying in some cases consisted in placing a thin, semi-opaque sheet of paper on the surface of the original image and sketching the contours. Read more

Visual Object Recognition in Datasets of Pre-modern Images
Read more
WIN-Kolleg: artistic and artifical Seeing

Das WIN-Projekt widmet sich Fragestellungen, die von den beiden Bildwissenschaften Computer Vision und Kunstgeschichte geteilt werden. Der gemeinsame Fokus ist das Bildverstehen als erkenntnistheoretisches und anwendungsorientiertes Problem. Read more

COMPOSITO: Artistic and Artifical Seeing

The interdisciplinary Frontier project connects computer vision with the field of art history and analyses early modern architecture with the help of machine learning and image processing. Read more

Objectrecognition and Automatic Comparision of the Sachsenspiegel Codices

Cultural heritage consists not only of innovations but also of their reproductions and variations. Therefore it is crucial to evaluate the quality of these reproductions of art as well as their stylistic and semantic changes. Read more