معرفی
Dr. Eric Brachmann is a Researcher at Heidelberg University's Visual Learning Lab (since 2017) and a Guest at Leibniz University Hannover (since 2019). He earned his Dr. rer. nat. in 2018 from TU Dresden (summa cum laude), preceded by a Diplom in media computer science (2012) and studies (2006–2012) at TU Dresden.
- Doctorate: TU Dresden (2018, summa cum laude)
- Diplom: TU Dresden (2012, passed with distinction)
- Education: Media and computer science (2006–2012)
His research focuses on Computer Vision and Machine Learning, particularly 6D object pose estimation, camera localization, and neural-guided optimization. His work bridges classical geometric methods (e.g., RANSAC) with modern deep learning techniques, including differentiable optimization and reinforcement learning for pose estimation.
His publications emphasize end-to-end learning, robust model fitting, and RGB-D image analysis, with applications in robotics and 3D scene understanding. Key contributions include DSAC, CONSAC, and neural extensions of RANSAC for efficient hypothesis sampling.
- 2018: GI Dissertation Award nomination
- 2014: ACCV Honorable Mention Demo Award
- 2012: Enno Heidebroek Award for top graduate
- 2008–2012: German National Academic Foundation scholarship
- 2008: IBM Award for intermediate diploma
As a co-organizer of ICCV and ECCV workshops, Eric drives collaboration in visual localization and 6D pose estimation. He has reviewed for CVPR, ICCV, NeurIPS, and TPAMI, earning recognition as an Outstanding Reviewer (CVPR 19, NeurIPS 19).
He has held industry roles at IBM (2010–2011) and T-Systems (2008–2009). At TU Dresden and Heidelberg, he taught courses on computer vision and 3D world reconstruction, supervised theses, and developed practical seminars.
