A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems

Autor: Jan Lellmann, Jörg Hendrik Kappes, Carsten Rother, Bogdan Savchynskyy, Bernhard X. Kausler, Nikos Komodakis, Bjoern Andres, Christoph Schnörr, Thorben Kröger, Sungwoong Kim, Sebastian Nowozin, Fred A. Hamprecht, Dhruv Batra
Přispěvatelé: Heidelberg University, Max Planck Institute for Informatics [Saarbrücken], Microsoft Research [Cambridge] (Microsoft), Microsoft Research, Virginia Tech [Blacksburg], Qualcomm Research Korea, University of Cambridge [UK] (CAM), Laboratoire d'Informatique Gaspard-Monge (LIGM), Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM), Technische Universität Dresden = Dresden University of Technology (TU Dresden)
Jazyk: angličtina
Rok vydání: 2015
Předmět:
FOS: Computer and information sciences
Computer science
Computer Vision and Pattern Recognition (cs.CV)
Computer Science - Computer Vision and Pattern Recognition
Inference
0102 computer and information sciences
02 engineering and technology
Energy minimization
Machine learning
computer.software_genre
01 natural sciences
Artificial Intelligence
0202 electrical engineering
electronic engineering
information engineering
Integer programming
Random field
Markov chain
business.industry
[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
010201 computation theory & mathematics
Pattern recognition (psychology)
Benchmark (computing)
Combinatorial optimization
020201 artificial intelligence & image processing
Computer Vision and Pattern Recognition
Artificial intelligence
business
computer
Software
Zdroj: International Journal of Computer Vision
International Journal of Computer Vision, Springer Verlag, 2015, ⟨10.1007/s11263-015-0809-x⟩
ISSN: 0920-5691
1573-1405
Popis: International audience; Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov Random Fields (MRF). This study provided valuable insights in choosing the best optimization technique for certain classes of problems. While these insights remain generally useful today, the phenomenal success of random field models means that the kinds of inference problems that have to be solved changed significantly. Specifically , the models today often include higher order interactions, flexible connectivity structures, large label-spaces of different car-dinalities, or learned energy tables. To reflect these changes, we provide a modernized and enlarged study. We present an empirical comparison of more than 27 state-of-the-art optimization techniques on a corpus of 2,453 energy minimization instances from diverse applications in computer vision. To ensure reproducibility, we evaluate all methods in the OpenGM 2 framework and report extensive results regarding runtime and solution quality. Key insights from our study agree with the results of Szeliski et al. for the types of models they studied. However, on new and challenging types of models our findings disagree and suggest that polyhedral methods and integer programming solvers are competitive in terms of runtime and solution quality over a large range of model types.
Databáze: OpenAIRE
Externí odkaz: