Software

Software

This page provides source code and other related files for several of our publications. For the license terms governing these (software) packages, please see the contents of the respective package. Generally, the software is for non-commercial personal and research use only. Please contact us, should you wish to use the software for commercial purposes. Also note that the software is generally provided as is, i.e. without any warranties.

Software Package Publication Venue
UnFlow (Unsupervised Networks for Optical Flow) AAAI 2018
ProbFlow ICCV 2017
Playing for Data ECCV 2016
Stereo Video Deblurring ECCV 2016
Discrete-Continuous Multi-Target Tracking PAMI (2015), CVPR 2013, CVPR 2012
Blind Deblurring with Interleaved RTFs WACV 2015
Piecewise Rigid 3D Scene Flow IJCV (2015), ECCV 2014, ICCV 2013
Shrinkage Fields CVPR 2014
Continuous Multi-Target Tracking PAMI (2014), ICCVWS 2011
Quantitative Analysis of Current Practices in Optical Flow IJCV (2014), CVPR 2010
Discriminative Non-Blind Deblurring CVPR 2013
An Evaluation of Data Costs for Optical Flow GCPR 2013
Learned Rotation-Invariant/Equivariant Models & Features CVPR 2012
Non-Blind Deblurring & Blind Denoising with Noise Estimation CVPR 2011
Sampling-based Learning & Inference for Pariwise & High-order MRFs CVPR 2010
Pictorial Structures for People Detection & Pose Estimation CVPR 2009
Fields of Experts IJCV 2009, CVPR 2005
 

UnFlow (Unsupervised Networks for Optical Flow, AAAI 2018)

This package provides source code for unsupervised learning of deep neural networks for optical flow estimation.
Relevant citation
(please cite this paper if you are using the software)
S. Meister, J. Hur, S. Roth, “UnFlow: Unsupervised Learning of Optical Flow with a Bidirectional Census Loss,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), Feb. 2018, to appear.
Source will be made available shortly
License see source code
Contact Simon Meister

ProbFlow (ICCV 2017)

This package provides source code for the joint estimation of optical flow and uncertainties.
Relevant citation
(please cite this paper if you are using the software)
A. S. Wannenwetsch, M. Keuper, S. Roth, “ProbFlow: Joint Optical Flow and Uncertainty Estimation,” in Proc. of the Sixteenth IEEE International Conference on Computer Vision (ICCV), Venice, Italy, Oct. 2017.
Source Bitbucket
License see source code
Contact Anne Wannenwetsch

Playing for Data (ECCV 2016)

This package provides source code for extracting synthetic data from computer games.
Relevant citation
(please cite this paper if you are using the software/dataset)
S. R. Richter, V. Vineet, S. Roth, and V. Koltun, “Playing for data: Ground truth from computer games,” in Proc. of the European Conference on Computer Vision (ECCV), J. Matas, B. Leibe, M. Welling and N. Sebe, Eds., ser. LNCS, Springer, 2016
Source Bitbucket
License see source code
Contact Stephan Richter

Stereo Video Deblurring (ECCV 2016)

This package provides source code for deblurring stereo videos.
Relevant citation
(please cite this paper if you are using the software)
A. Sellent, C. Rother, and S. Roth, “Stereo video deblurring,” in Proc. of the European Conference on Computer Vision (ECCV), B. Leibe, J. Matas, N. Sebe and M. Welling, Eds., ser. LNCS, vol. 9906, Springer, 2016, pp. 558–575.
Source code package
License see source code
Contact Anita Sellent

Discrete-Continuous Multi-Target Tracking (CVPR 2012, CVPR 2013, PAMI 2015)

This package provides source code for our work on discrete-continuous energy minimization for multi-target tracking.
Relevant citation
(please cite this paper if you are using the source code)
A. Milan, K. Schindler, S. Roth, “Multi-target tracking by discrete-continuous energy minimization,” IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), to appear.
Other relevant papers A. Milan, S. Roth, and K. Schindler, “Detection- and trajectory-level exclusion in multiple object tracking,”in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Portland, Oregon, Jun. 2013, pp. 3682–3689.

A. Andriyenko, K. Schindler, and S. Roth, “Discrete-continuous optimization for multi-target tracking,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, Rhode Island, Jun. 2012, pp. 1926–1933.
Source Bitbucket
License see source code
Contact Anton Milan

Blind Deblurring with Interleaved RTFs (WACV 2015)

This package provides source code for our work on blind deblurring by interleaving kernel estimation with discriminative deblurring using RTFs.
Relevant citation
(please cite this paper if you are using the source code)
K. Schelten, S. Nowozin, J. Jancsary, C. Rother, and S. Roth, “Interleaved regression tree field cascades for blind image deconvolution,” in IEEE Winter Conference on Applications of Computer Vision (WACV), Waikoloa Beach, HI, Jan. 2015, pp. 494-501.
Source Bitbucket
License see source code
Contact

Piecewise Rigid 3D Scene Flow (ICCV 2013, ECCV 2014, IJCV 2015)

This package provides source code for our work on 3D scene flow estimation with a piecewise rigid scene representation.
Relevant citation
(please cite this paper if you are using the source code)
C. Vogel, K. Schindler, and S. Roth, “3D scene flow estimation with a piecewise rigid scene model,” International Journal of Computer Vision (IJCV), vol. 111, no. 3, 2015.
Other relevant papers C. Vogel, K. Schindler, and S. Roth, “Piecewise rigid scene flow,” in Proc. of the IEEE International Conference on Computer Vision (ICCV), Sydney, Australia, Dec. 2013, pp. 1377–1384

C. Vogel, S. Roth, and K. Schindler, “View-consistent 3D scene flow estimation over multiple frames,” in Proc. of the European Conference on Computer Vision (ECCV), D. Fleet, T. Pajdla, B. Schiele, and T. Tuytelaars, Eds., ser. LNCS, vol. 8692, Springer, 2014, pp. 263–278.
Source GitHub
License see source code
Contact Christoph Vogel

Shrinkage Fields (CVPR 2014)

This package provides source code for our work on discriminative shrinkage field models for efficient high-quality image restoration (denoising & non-blind deblurring).
Relevant citation
(please cite this paper if you are using the source code)
U. Schmidt and S. Roth, “Shrinkage fields for effective image restoration,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, Ohio, Jun. 2014, pp. 2774–2781.
Source Version 1.1 (November 19, 2014), GitHub
License see source code
Contact Uwe Schmidt

Continuous Multi-Target Tracking (ICCVWS 2011, PAMI 2014)

This package provides source code for our work on continuous energy minimization for multi-target tracking.
Relevant citation
(please cite this paper if you are using the source code)
A. Milan, S. Roth, and K. Schindler, “Continuous energy minimization for multi-target tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), vol. 36, no. 1, pp. 58–72, Jan. 2014.
Other relevant papers A. Andriyenko, S. Roth, and K. Schindler, “An analytical formulation of global occlusion reasoning for multi-target tracking,” in 11th International IEEE Workshop on Visual Surveillance, Barcelona, Spain, Nov. 2011, pp. 1839–1846.

A. Andriyenko and K. Schindler, Multi-target Tracking by Continuous Energy Minimization” in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Colorado Springs, Colorado, June 2011.
Source Bitbucket
License see source code
Contact Anton Milan

Quantitative Analysis of Current Practices in Optical Flow (CVPR 2010, IJCV 2014)

This package provides source code for our work on quantitatively analyzing current practices in optical flow estimation algorithms, a.k.a. “the secrets of optical flow”.
Relevant citation
(please cite this paper if you are using the source code)
D. Sun, S. Roth, and M. J. Black, “A quantitative analysis of current practices in optical flow estimation and the principles behind them,” International Journal of Computer Vision (IJCV), vol. 106, no. 2, pp. 115–137, Jan. 2014.
Other relevant papers D. Sun, S. Roth, and M. J. Black, “Secrets of optical flow estimation and their principles,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Francisco, California, Jun. 2010, pp. 2432–2439.
Source IJCV code, CVPR 2010 code
License see source code
Contact Deqing Sun

Discriminative Non-Blind Deblurring (CVPR 2013)

This package provides source code for our work on discriminative models for non-blind image deblurring.
Relevant citation
(please cite this paper if you are using the source code)
U. Schmidt, C. Rother, S. Nowozin, J. Jancsary, and S. Roth, “Discriminative non-blind deblurring,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Portland, Oregon, Jun. 2013, pp. 604–611.
Source Version 1.1 (May 8, 2014)
License see source code
Contact Uwe Schmidt

An Evaluation of Data Costs for Optical Flow (GCPR 2013)

This package provides source code for our evaluation study of different data costs for optical flow estimation.
Relevant citation
(please cite this paper if you are using the source code)
C. Vogel, K. Schindler, and S. Roth, “An evaluation of data costs for optical flow,” in Proc. of the German Conference on Pattern Recognition (GCPR), J. Weickert, M. Hein, and B. Schiele, Eds., ser. LNCS, vol. 8142, Springer, 2013, pp. 343–353.
Source GitHub
License see source code
Contact Christoph Vogel

Learned Rotation-equivariant/invariant Models & Features (CVPR 2012)

This package provides source code for our work on learning rotation-equivariant and rotation-invariant models and features for restoration and classification.
Relevant citation
(please cite this paper if you are using the source code)
U. Schmidt and S. Roth, “Learning rotation-aware features: From invariant priors to equivariant descriptors,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, Rhode Island, Jun. 2012, pp. 2050–2057.
Source Version 1.0 (August 2, 2014)
License see source code
Contact Uwe Schmidt

Non-Blind Deblurring & Blind Denoising with Noise Estimation (CVPR 2011)

This package provides source code for our work on non-blind Bayesian deblurring as well as blind denoising with integrated noise estimation.
Relevant citation
(please cite this paper if you are using the source code)
U. Schmidt, K. Schelten, and S. Roth, “Bayesian deblurring with integrated noise estimation,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Colorado Springs, Colorado, Jun. 2011, pp. 2625–2632.
Source Version 1.0 (June 17, 2011)
License see source code
Contact Uwe Schmidt

Sampling-based Learning & Inference for Pairwise & High-Order MRFs (CVPR 2010)

This package provides source code for our work on sampling methods for performing learning and (MMSE) inference in pairwise and high-order MRFs.
It provides a MATLAB implementation of pairwise MRFs as well as Fields of Experts, each based on flexible Gaussian scale mixture (GSM) potentials. The package also provides an efficient auxiliary variable Gibbs sampler for learning and inference along with demo code for model learning, evaluating the models' statistical properties, and for image restoration based on MMSE estimation.
Relevant citation
(please cite this paper if you are using the source code)
U. Schmidt, Q. Gao, and S. Roth, “A generative perspective on MRFs in low-level vision,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Francisco, California, Jun. 2010, pp. 1751–1758.
Source Version 1.1 (June 17, 2011)
License see source code
Contact Uwe Schmidt

Pictorial Structures for People Detection & Pose Estimation (CVPR 2009)

This package provides source code for our work on discriminative appearance models for pictorial structures for people detection and pose estimation.
Relevant citation
(please cite this paper if you are using the source code)
M. Andriluka, S. Roth, and B. Schiele, “Pictorial Structures Revisited: People Detection and Articulated Pose Estimation,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Miami, Florida, Jun. 2009, pp. 1014–1021.
Source project page
License see source code
Contact Micha Andriluka

Fields of Experts (IJCV 2009, CVPR 2005)

This package provides source code for denoising and image inpainting with Fields of Experts (FoE), as well as learned example models.
Relevant citation
(please cite this paper if you are using the source code)
S. Roth and M. J. Black, “Fields of experts,” International Journal of Computer Vision (IJCV), vol. 82, no. 2, pp. 205–229, Apr. 2009.
Other relevant papers S. Roth and M. J. Black, “Fields of experts: A framework for learning image priors,” in Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2, San Diego, California, Jun. 2005, pp. 860–867.
Note This code is provided for reproducible research only! For most purposes, you should refer to this more recent implementation.
Source separate page
License see source code
Contact Stefan Roth