Search

Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov random fields. 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 cardinalities, 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 2453 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. [ABSTRACT FROM AUTHOR]

Nanoscale time-dependent mechanical-electrical coupled behavior of single crystal ZnO nanorods was systematically explored, which is essential for accessing the long-term reliability of the ZnO nanorod-based flexible devices. A series of compression creep tests combined with in-situ electrical measurement was performed on vertically-grown single crystal ZnO nanorods. Continuous measurement of the current (I)-voltage (V) curves before, during, after the creep tests revealed that I is non-negligibly increased as a result of the time-dependent deformation. Analysis of the I-V curves based on the thermionic emission-diffusion theory allowed extraction of nanorod resistance, which was shown to decrease as time-dependent deformation. Finally, based on the observations in this study, a simple analytical model for predicting the reduction in nanorod resistance as a function of creep strain that is induced from diffusional mechanisms is proposed, and this model was demonstrated to be in an excellent agreement with the experimental results. [ABSTRACT FROM AUTHOR]