Rounding-based Moves for Semi-Metric Labeling
Authors: M. Pawan Kumar, Puneet K. Dokania
JMLR 2016 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the efficacy of rounding-based moves by comparing them to several state of the art methods using both synthetic and real data. ... Figure 1 shows the results for the first data set. ... Figure 2 shows the results for the second data set. ... Figure 3-Figure 8 shows the results for various standard pairs of images. |
| Researcher Affiliation | Academia | M. Pawan Kumar EMAIL Department of Engineering Science University of Oxford Parks Road OX1 3PJ United Kingdom Puneet K. Dokania EMAIL Center for Visual Computing Centrale Supelec 92295 Chˆatenay-Malabry France |
| Pseudocode | Yes | Algorithm 1 The complete rounding procedure. Algorithm 2 The complete move-making algorithm. Algorithm 3 The interval rounding procedure. Algorithm 4 The interval move-making algorithm. Algorithm 5 The hierarchical rounding procedure. Algorithm 6 The hierarchical move-making algorithm. |
| Open Source Code | No | The paper does not contain any explicit statements or links indicating that the authors' source code for the described methodology is publicly available. |
| Open Datasets | Yes | For the above problem, we use the unary potentials and edge weights that are specified by Szeliski et al. (2008). ... Figure 3-Figure 8 shows the results for various standard pairs of images. |
| Dataset Splits | No | The paper describes how synthetic datasets were generated and used in experiments, but it does not specify any training, validation, or test splits. For the dense stereo correspondence, it mentions using images but provides no information on dataset splits. |
| Hardware Specification | No | The paper does not explicitly describe the specific hardware (e.g., GPU models, CPU types, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions using 'minimum st-cut solvers' and 'TRW-S (Kolmogorov, 2006)' but does not provide specific version numbers for these or any other software libraries or frameworks used in their implementation. |
| Experiment Setup | Yes | The unary potentials were sampled from a uniform distribution over [0, 10]. The edge weights were sampled from a uniform distribution over [0, 3]. ... For the truncated linear and truncated quadratic cases, we present results of interval move-making (INT) using the optimal interval length reported by Kumar and Torr (2008). ... We use two types of pairwise potentials: (i) truncated linear with the truncation set at 4; and (ii) truncated quadratic with the truncation set at 16. |