Simon Doll

Machine Learning Engineer

Scene Understanding Group

Mercedes-Benz R&D

Hey there! I am Simon, a PhD student with the Scene Understanding Group at Mercedes-Benz R&D, as well as the University of Tübingen where I am advised by Prof. Hendrik P.A. Lensch.

My research interests lie in machine learning, especially at the intersection between computer vision and robotics. In particular, I am interested in incorporating physical knowledge to learned representations as well as finding robust and scalable representations for multimodal perception.

Additionally, I founded two startups with my twin brother in healthcare where we focused on machine learning methods for vision centric applications in dermatology as well as improved patient anamnesis and doctor letter generation.

Besides research I am a passionate musician, love surfing and I am always in for a good cup of Coffee, Espresso - or anything else with caffeine.

selected publications

SpatialDETR: Robust Scalable Transformer-Based 3D Object Detection From Multi-view Camera Images With Global Cross-Sensor Attention

Simon Doll, Richard Schulz, Lukas Schneider, and 3 more authors

In Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXXIX, 2022

Bib Page PDF Video Code

@inproceedings{doll2022spatialdetr,
  title = {SpatialDETR: Robust Scalable Transformer-Based 3D Object Detection From Multi-view Camera Images With Global Cross-Sensor Attention},
  author = {Doll, Simon and Schulz, Richard and Schneider, Lukas and Benzin, Viviane and Enzweiler, Markus and Lensch, Hendrik PA},
  booktitle = {Computer Vision--ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23--27, 2022, Proceedings, Part XXXIX},
  pages = {230--245},
  year = {2022},
  organization = {Springer},
}

Star-track: Latent motion models for end-to-end 3d object tracking with adaptive spatio-temporal appearance representations

Simon Doll, Niklas Hanselmann, Lukas Schneider, and 3 more authors

IEEE Robotics and Automation Letters, 2023

Bib Page PDF Video

@article{doll2022startrack,
  title = {Star-track: Latent motion models for end-to-end 3d object tracking with adaptive spatio-temporal appearance representations},
  author = {Doll, Simon and Hanselmann, Niklas and Schneider, Lukas and Schulz, Richard and Enzweiler, Markus and Lensch, Hendrik PA},
  journal = {IEEE Robotics and Automation Letters},
  volume = {9},
  number = {2},
  pages = {1326--1333},
  year = {2023},
  publisher = {IEEE},
}

Dualad: Disentangling the dynamic and static world for end-to-end driving

Simon Doll, Niklas Hanselmann, Lukas Schneider, and 4 more authors

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

Bib Page PDF Video

@inproceedings{doll2024dualad,
  title = {Dualad: Disentangling the dynamic and static world for end-to-end driving},
  author = {Doll, Simon and Hanselmann, Niklas and Schneider, Lukas and Schulz, Richard and Cordts, Marius and Enzweiler, Markus and Lensch, Hendrik},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages = {14728--14737},
  year = {2024},
}