We are building machine learning models for computer vision, natural language and robotics. In particular, we focus on learning 2D and 3D representations of objects and scenes, reconstructing geometry and materials and learning discriminative and generative models. We also investigate how complex knowledge can be incorporated into machine learning algorithms for making them robust to variations in our complex world. Applications include self-driving cars, household robots, virtual/augmented reality and scientific document analysis. You can follow us on Google Scholar, YouTube, Twitter, Facebook and on our Blog. Pictures from group activities can be found here.
- Best Paper Award: CVPR 2021, 3DV 2017, 3DV 2015, GCPR 2015
- Best Paper Finalist: CVPR 2019 (2x), CVPR 2013
- Most influential CVPR papers: DVR #15 in 2020, Occupancy Nets #14 in 2019, KITTI #1 in 2012 (source)
- Awards: Longuet-Higgins Prize 2022, Mark Everingham Price 2021, Teaching Award 2021, ERC Starting Grant 2019, IEEE PAMI Young Researcher Award 2018, Heinz Maier-Leibnitz Prize 2017, German Pattern Recognition Prize 2017