Philipp Krähenbühl 
Department of Computer Science
University of Texas at Austin
2317 Speedway
Austin, TX 78712-1757

email: philkr (at)


I am an Assistant Professor in the Department of Computer Science at the University of Texas at Austin. I received my PhD in 2014 from the CS Department at Stanford University and then spent two wonderful years as a PostDoc at UC Berkeley.

My research interests lie in Computer Vision, Machine learning and Computer Graphics. I'm particularly interested in deep learning, as well as image segmentation and understanding.


Learning Dense Correspondence via 3D-guided Cycle Consistency
Tinghui Zhou, Philipp Krähenbühl, Mathieu Aubry, Qixing Huang and Alyosha Efros
CVPR 2016
[PDF] [Details] [Code coming soon]
Context Encoders: Feature Learning by Inpainting
Deepak Pathak, Philipp Krähenbühl, Jeff Donahue, Trevor Darrell and Alyosha Efros
CVPR 2016
[PDF] [Details] [Code]
Data-dependent initializations of convolutional neural networks
Philipp Krähenbühl, Carl Doersch, Jeff Donahue and Trevor Darrell
ICLR 2016
[PDF] [Details and code] [py-faster-rcnn training scripts]
Constrained Convolutional Neural Networks for Weakly Supervised Segmentation
Deepak Pathak, Philipp Krähenbühl and Trevor Darrell
ICCV 2015
[PDF] [Supplement] [Slides] [Details] [Code]
Learning Data-driven Reflectance Priors for Intrinsic Image Decomposition
Tinghui Zhou, Philipp Krähenbühl and Alyosha Efros
ICCV 2015
[PDF] [Details] [Slides] [Code]
Learning a Discriminative Model for the Perception of Realism in Composite Images
Jun-Yan Zhu, Philipp Krähenbühl, Eli Shechtman and Alyosha Efros
ICCV 2015
[PDF] [Details] [Slides] [Code]
Learning to propose objects
Philipp Krähenbühl and Vladlen Koltun
CVPR 2015
[PDF] [Details] [Code]
Geodesic Object Proposals
Philipp Krähenbühl and Vladlen Koltun
ECCV 2014 (oral)
[PDF] [Details] [Slides] [Poster] [Code] [Data]

Parameter Learning and Convergent Inference for Dense Random Fields
Philipp Krähenbühl and Vladlen Koltun
ICML 2013
[PDF] [Details] [Project Page] [Code]
Efficient Nonlocal regularization for Optical Flow
Philipp Krähenbühl and Vladlen Koltun
ECCV 2012
[PDF] [Details]
Saliency Filters: Contrast Based Filtering for Salient Region Detection
Federico Perazzi, Philipp Krähenbühl, Yael Pritch and Alexander Hornung
CVPR 2012
[PDF] [Project page] [Code]
Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
Philipp Krähenbühl and Vladlen Koltun
NIPS 2011 (best student paper)
[PDF] [Details] [Project Page] [Code]
  Gesture Controllers
Sergey Levine, Philipp Krähenbühl, Sebastian Thrun, Vladlen Koltun
Proceedings of ACM SIGGRAPH 2010 (Los Angeles, July 25-29, 2010), ACM Transactions on Graphics
Retargeting of Streaming Video
Philipp Krähenbühl, Manuel Lang, Alexander Hornung and Markus Gross
Proceedings of ACM SIGGRAPH Asia (Yokohama, Japan, December 16-19, 2009), ACM Transactions on Graphics
Art Directable Retargeting for Streaming Video
Philipp Krähenbühl
Bachelor Thesis, ETH Zurich, February 2009


About my last name

I'm well aware that my last name is not the easiest one to write or cite (and I saw it butchered a bunch of times over the years). So to make things easier just pick your document type below and copy the string:

Regular text

Latex & Bibtex


If all the above fail, just use "Kraehenbuehl".