Robert-Jan Bruintjes

My interests are in integrating prior knowledge into Convolutional Neural Networks to benefit data efficiency and/or robustness. Topics of choice include group equivariant CNNs and attempts to reduce texture bias in CNNs.

Prior Experience

  • R&D Lead

    Sightcorp, 2017 - 2019

Education

  • MSc in Artificial Intelligence, 2017

    University of Amsterdam

  • BSc in Computer Science, 2014

    Utrecht University

Recent

Lecture Self-Attention

I gave a lecture on self-attention for Computer Vision. Slides.

VIPriors challenges 2020 report

We published a report on the VIPriors 2020 challenges.

ECCV Workshop

We organized the Visual Inductive Priors workshop.

More updates

Publications

Probabilistic multileave for online retrieval evaluation

Experience

 
 
 
 
 

PhD candidate

TU Delft

Sep 2019 – Present Delft, the Netherlands
 
 
 
 
 

R&D Lead

Sightcorp

Sep 2017 – Aug 2019 Amsterdam, the Netherlands
At Sightcorp I applied state-of-the-art Deep Learning for Computer Vision while constrained by business priorities and maintaining a solid code base for repeatable experimentation. I lead my colleagues and myself to achieve competitive results for face detection systems by exploiting overlapping expertise and coming up with innovative solutions by collaborating in brainstorming.

Contact

I am always willing to talk about collaboration or supervision. Please find my contact info at the top of the page.

Office

TU Delft, building 28, room 5.W.600
Van Mourik Broekmanweg 6
2628 XE Delft
The Netherlands