About me

Hi, I am Steffen Wolf, have obtained my PhD in the image analysis and learning group of Prof. Fred Hamprecht and have started my postdoc at MRC Laboratory of Molecular Biology in Cambridge, in the group of Kate McDole, where am working on Machine Learning and Computer Vision methods for image analysis.

In particular, I am interested in understanding self- and unsupervised learning methods for training neural networks efficiently and developing complementary (segmentation) algorithms that synergize with deep learning. The applications of my research aim to contribute to a better understanding of life sciences by aiding in the quantitative analysis, most importantly in the application of image segmentation of microscopy images.

I developed the Mutex Watershed (MWS), an efficient segmentation algorithm and analyzed its mathematical foundation. It utilizes deep neural that predict long-range affinities and scales to large images/volumes. If you have Python experience and want to work with the MWS yourself, check out our affogato library and tutorials. If you are interested in applying our pipeline on your data there is good news. We have acquired a software development grant and are making the Mutex Watershed and interactive network training available as a napari plugin.

Publications:

Designing Mosaic Cells for Self-Supervised Learning

Designing Mosaic Cells for Self-Supervised Learning

Steffen Wolf, Jan Funke and Katie McDole
Unsupervised Learning of Spatial Embeddings for Cell Segmentation

Unsupervised Learning of Spatial Embeddings for Cell Segmentation

Steffen Wolf, Katie McDole and Jan Funke
Inpainting Networks Learn to Separate Cells in Microscopy Images

Inpainting Networks Learn to Separate Cells in Microscopy Images

Steffen Wolf, Fred Hamprecht and Jan Funke
Learning the Arrow of Time for Problems in Reinforcement Learning

Learning the Arrow of Time for Problems in Reinforcement Learning

Nasim Rahaman, Steffen Wolf, Anirudh Goyal, Roman Remme, Yoshua Bengio
The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning

The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning

Steffen Wolf, Alberto Bailoni, Constantin Pape, Nasim Rahaman, Anna Kreshuk, Ullrich K├Âthe, and Fred A. Hamprecht

Projects