Welcome to the Computer Vision Lab Dresden

Computer Vision is a science that develops models and methods for understanding, analyzing, acquiring and processing images and generally high-dimensional structured data. Computer Vision is an inter-disciplinary field with strong links to Machine Learning, Optimization, Biology, Computer Graphics, and Human Computer Interaction. The mission of the Computer Vision Lab Dresden is to develop novel theoretical concepts which are practically relevant. We work on a broad range of application areas, from image matching, via semantic scene understanding, to Bio Imaging. On the theoretical side we focus on optimization and learning in probabilistic graphical models.

Latest News

  • MSF Workshop at ICCV 2015

    MSF Workshop at ICCV 2015

    CVLD organizes together with Rama Chellappa, Christian Heipke, Alper Yilmaz, Clément Mallet, Yury Vizilter the 2nd ICCV2015 Workshop: Multi-Sensor Fusion for Dynamic Scene Understanding (MSF 2015), …
  • 3 papers accepted to ICCV 2015

    3 papers accepted to ICCV 2015

    (1) Learning to compare a synthetic rendered image with a real observation Image for better 6D Object Pose estimation (2) …
  • 6D Object Pose  Workshop at ICCV2015

    6D Object Pose Workshop at ICCV2015

    We are helping to organize the 1st International Workshop on Recovering 6D Object Pose In conjuction with ICCV 2015, Santiago. …
  • Inverse Rendering Workshop at ICCV2015

    Inverse Rendering Workshop at ICCV2015

    CVLD organizes together with Stefan Roth and Peter Gehler a full day workshop at ICCV2015. Look here to find out …
  • ICCV 2015 Pose Challenges are online

    ICCV 2015 Pose Challenges are online

        Our ICCV 2015 Pose Challenges – Articulated Object Challenge & Occluded Object Challenge – are online.


The Rich Scene Model (ERC Consolidator Grant)

Given a sequence of images the goal is to recover a rich, detailed representation of the 3D world, ranging from physical to semantical aspects. To achieve this we investigate new ways to combine feature learning, modelling, physical laws, and optimization in large-scale discrete-continuous-valued probabilistic graphical model. 

Our Research

Optimization & Learning

with a focus on graphical models, deep models, and large-scale optimization.

Image Matching

with a focus on 3-9DoF scene flow, as well as jointly recovering multiple physical aspects.

Scene Understanding

with a focus on 3D semantic scene under-standing, and model-based vision.

Bio Imaging

with a focus on segmentation, tracking, and matching – jointly with teams from MPI-CBG.

Image Analysis

with a focus on segmentation, interactive techniques, as well as camera design.

Regular Events

Collaborators & Industrial Partners

We have strong international collaborations with Oxford University, University College London, Skoltech Moscow, Stanford University, HCI Heidelberg, TU Darmstadt, Microsoft Research Cambridge, Prague University and others.
We also have industrial collaborations with Daimler Research, Adobe (Seattle) and Basler AG.

Office Hours

Katrin Heber

Monday, Tuesday, Wednesday, Friday
14:00 - 16:00

Prof. PhD. Carsten Rother

Head of CVLD.
Please arrange an appointment with Katrin Heber (Secretary).