Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Paul Swoboda

Paul Swoboda

ClusterFuG: Clustering Fully connected Graphs by Multicut

Jan 28, 2023
Ahmed Abbas, Paul Swoboda

Figure 1 for ClusterFuG: Clustering Fully connected Graphs by Multicut

Figure 2 for ClusterFuG: Clustering Fully connected Graphs by Multicut

Figure 3 for ClusterFuG: Clustering Fully connected Graphs by Multicut

Figure 4 for ClusterFuG: Clustering Fully connected Graphs by Multicut

Access Paper or Ask Questions

DOGE-Train: Discrete Optimization on GPU with End-to-end Training

May 23, 2022
Ahmed Abbas, Paul Swoboda

Figure 1 for DOGE-Train: Discrete Optimization on GPU with End-to-end Training

Figure 2 for DOGE-Train: Discrete Optimization on GPU with End-to-end Training

Figure 3 for DOGE-Train: Discrete Optimization on GPU with End-to-end Training

Figure 4 for DOGE-Train: Discrete Optimization on GPU with End-to-end Training

* Alert before printing: pg. 18-25 only contain per instance results, can possibly be skipped

Access Paper or Ask Questions

FastDOG: Fast Discrete Optimization on GPU

Nov 19, 2021
Ahmed Abbas, Paul Swoboda

Figure 1 for FastDOG: Fast Discrete Optimization on GPU

Figure 2 for FastDOG: Fast Discrete Optimization on GPU

Figure 3 for FastDOG: Fast Discrete Optimization on GPU

Figure 4 for FastDOG: Fast Discrete Optimization on GPU

* Alert before printing: last 10 pages just contains detailed results can possibly be skipped

Access Paper or Ask Questions

RAMA: A Rapid Multicut Algorithm on GPU

Sep 04, 2021
Ahmed Abbas, Paul Swoboda

Figure 1 for RAMA: A Rapid Multicut Algorithm on GPU

Figure 2 for RAMA: A Rapid Multicut Algorithm on GPU

Figure 3 for RAMA: A Rapid Multicut Algorithm on GPU

Figure 4 for RAMA: A Rapid Multicut Algorithm on GPU

Access Paper or Ask Questions

Making Higher Order MOT Scalable: An Efficient Approximate Solver for Lifted Disjoint Paths

Aug 24, 2021
Andrea Hornakova, Timo Kaiser, Paul Swoboda, Michal Rolinek, Bodo Rosenhahn, Roberto Henschel

Figure 1 for Making Higher Order MOT Scalable: An Efficient Approximate Solver for Lifted Disjoint Paths

Figure 2 for Making Higher Order MOT Scalable: An Efficient Approximate Solver for Lifted Disjoint Paths

Figure 3 for Making Higher Order MOT Scalable: An Efficient Approximate Solver for Lifted Disjoint Paths

Figure 4 for Making Higher Order MOT Scalable: An Efficient Approximate Solver for Lifted Disjoint Paths

* ICCV 2021. Short version published at CVPR 2021 RVSU workshop https://omnomnom.vision.rwth-aachen.de/data/RobMOTS/workshop/papers/9/CameraReady/paper_V3.pdf . Implementation available at https://github.com/LPMP/LPMP and https://github.com/TimoK93/ApLift

Access Paper or Ask Questions