Hyperplane Clustering via Dual Principal Component Pursuit

Authors: Manolis C. Tsakiris, René Vidal

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on synthetic and real data, are shown to outperform or be competitive to the state-of-the-art. 5. Experimental Evaluation.
Researcher Affiliation Academia Manolis C. Tsakiris 1 Ren e Vidal 1 1Center for Imaging Science, Johns Hopkins University, Baltimore, MD, USA.
Pseudocode No The paper describes algorithms in prose, but does not include structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes In this section we explore various hyperplane clustering algorithms using the benchmark dataset NYUdepth V2 (Silberman et al., 2012).
Dataset Splits No The paper mentions evaluating algorithms on 'manually annotated 92 of the 1449 scenes' from the NYUdepth V2 dataset, but does not specify train, validation, or test dataset splits.
Hardware Specification Yes a MATLAB implementation on a standard Mac Book Pro with a dual core 2.5GHz processor and a total of 4GB cache memory
Software Dependencies No The paper mentions 'a MATLAB implementation' and 'an optimized LP solver such as GUROBI' but does not specify version numbers for these software dependencies.
Experiment Setup Yes DPCP-r, which uses a maximum of 20 iterations in (3), while REAPER and DPCP-IRLS use a maximum of 100 iterations and convergence accuracy 10 3. We use 10 independent restarts.