Till Kroeger, PhD Student

Welcome! My name is Till Kroeger and I am a PhD candidate under the supervision of Prof. Luc Van Gool at the Computer Vision Laboratory of ETH Zurich. I am working in collaboration with the Varcity project.

In my work I focus on (1) multi-view tracking of multiple objects, (2) simultaneous localization and mapping, (3) fast image-based registration for videos and camera pose estimation, (4) fast optical flow and depth from stereo.

Before joining the Computer Vision Lab at ETH Zurich I completed two master's degrees in (1) Computer Vision and Computational Linguistics and (2) European Medieval History and Middle Eastern Studies at the University of Heidelberg, Germany.

If you need to contact me, you can find my contact details here. My CV is available here.

You can also find me on: Twitter, Google scholar, LinkedIn, Xing.

News

2016 Jul: Our paper "Fast Optical Flow using Dense Inverse Search" is accepted to ECCV 2016
2015 Jun: I joined the Max Planck ETH Center for Learning Systems as Associate (Doctoral) Fellow.
2015 Apr: The dataset for our paper "Joint Vanishing Point Extraction and Tracking" is available.
2015 Mar: Our papers "Joint Vanishing Point Extraction and Tracking" and "Metric Imitation by Manifold Transfer" are accepted to CVPR 2015.
2014 Sep: Code for our paper "Video Registration to SfM Models" is available.
2014 Jul: Our paper "Video Registration to SfM Models" is accepted to ECCV 2014.
2014 Jul: Our paper "Multi-View Tracking of Multiple Targets with Dynamic Cameras" is accepted to GCPR 2014.
2014 Jun: I am accepted for the Vision and Sports Summer School 2014 at Czech Technical University, Prague.
2012 Oct: I joined the the Computer Vision Laboratory, ETH Zurich in the Varcity project as a PhD student.


Publications
ECCV 2016

Fast Optical Flow using Dense Inverse Search
T. Kroeger, R. Timofte, D. Dai, L. Van Gool.
[Pdf] [Supplementary Material] [Project Page]

CVPR 2015

Joint Vanishing Point Extraction and Tracking
T. Kroeger, D. Dai, L. Van Gool. (Oral Presentation)
[Pdf] [Supplementary Material] [Project Page] [Video of Talk at CVPR] [Slides]

CVPR 2015

Metric Imitation by Manifold Transfer
D. Dai, T. Kroeger, R. Timofte, L. Van Gool.
[Pdf] [Supplementary Material] [Project Page]

WACV-W 2015

Discovery of Sets of Mutually Orthogonal Vanishing Points in Videos
T. Kroeger, D. Dai, R. Timofte, L. Van Gool.
[Pdf]

ECCV 2014

Video Registration to SfM Models
T. Kroeger, L. Van Gool.
[Pdf] [Supplementary Material] [Project Page]

GCPR 2014

Multi-View Tracking of Multiple Targets with Dynamic Cameras
T. Kroeger, R. Dragon, L. Van Gool.
[Pdf] [Video]


Datasets & Code
Dataset

Vanishing Point annotations for Antwerp Street-View Dataset

Dataset

Antwerp Street-View Dataset for Video Registration

Code

Video Registration Code to SfM Models

Code

Bundle Adjustment with Matlab interface

Code

Fast Optical Flow using Dense Inverse Search