Robotics and Biology Laboratory

Pia Bideau

Research Interests

I'm a postdoctoral researcher at TU Berlin and part of the Cluster Science of Intelligence as well as the RBO lab. I in particular like research at the intersection of Computer Vision, Robotics and Machine learning.

Before joining the Cluster of Intelligence, I did my PhD at University of Massachusetts, Amherst (USA) working with Prof. Erik Learned-Miller.  During my PhD I had the opportunity to work together with Cordelia Schmid and Karteek Alahari as part of an internship at Inria in Grenoble (France) what was a lot of fun. I got my M.Sc. in Electrical Engineering and Information Technology from Ruhr-University Bochum.

In my thesis I developed new methods for moving object segmentation in unconstrained videos. Motion is a key ability that we as living beings have to explore our environment. Our motion for example helps us to perceive depth, and the motion of objects helps us to recognize these objects even if those are unknown to us. In general my research aims at the following questions, how can we incorporate physical principles into vision systems? And how can we adopt learned principles to a never previously seen scenario? Creating this kind of common sense understanding for machines requires consideration of many different aspects (physics of the world, social interactions, cultural background, etc.). To me, especially this makes the intersection between computer vision and robotics an incredible challenging but also very interesting research area to explore.

Wiss. Mitarbeiter_in, PostDoc


Gebäude MAR
Raum MAR



Halawa, Marah; Hellwich, Olaf; Bideau, Pia
Action-based Contrastive Learning for Trajectory Prediction
Proceedings of the European Conference on Computer Vision (ECCV)
Oktober 2022
Bideau, Pia; Learned-Miller, Erik; Schmid, Cordelia; Alahari, Karteek
The Right Spin: Learning Object Motion from Rotation-Compensated Flow Fields
arXiv preprint arXiv:2203.00115


Gu, Cheng; Learned-Miller, Erik; Gallego, Guillermo; Bideau, Pia
The Spatio-Temporal Poisson Point Process: A Simple Model for the Alignment of Event Camera Data
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)
Oktober 2021


Tang, Zhipeng; Delattre, Fabien; Bideau, Pia; Corner, Mark D; Learned-Miller, Erik
C-14: assured timestamps for drone videos
Proceedings of the 26th Annual International Conference on Mobile Computing and Networking, Seite 1-13
Bideau, Pia
Motion Segmentation-Segmentation of Independently Moving Objects in Video


Bideau, Pia; Menon, Rakesh R; Learned-Miller, Erik
MoA-Net: self-supervised motion segmentation
Proceedings of the European Conference on Computer Vision (ECCV) Workshops,
Bideau, Pia; RoyChowdhury, Aruni; Menon, Rakesh R; Learned-Miller, Erik
The best of both worlds: Combining cnns and geometric constraints for hierarchical motion segmentation
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Seite 508-517


Bideau, Pia; Learned-Miller, Erik
A detailed rubric for motion segmentation
arXiv preprint arXiv:1610.10033
Bideau, Pia; Learned-Miller, Erik
It’s moving! A probabilistic model for causal motion segmentation in moving camera videos
European Conference on Computer Vision, Seite 433-449