Homomorphic Sensing
Authors: Manolis Tsakiris, Liangzu Peng
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | On the algorithmic level we exhibit two dynamic programming based algorithms, which to the best of our knowledge are the first working solutions for the unlabeled sensing problem for small dimensions. One of them, additionally based on branch-and-bound, when applied to image registration under affine transformations, performs on par with or outperforms state-of-the-art methods on benchmark datasets. |
| Researcher Affiliation | Academia | 1School of Information Science and Technology, Shanghai Tech University, Shanghai, China. |
| Pseudocode | No | The paper describes algorithms in prose, but it does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing its source code or a link to a code repository for the methodology described. |
| Open Datasets | Yes | We compare 1) Algorithm-C with state-of-the-art image registration techniques... using a subset of the benchmark datasets used by Lian et al. 2017. Image registration using the synthetic benchmark dataset The Chinese Character (Chui & Rangarajan, 2003) (2a-2c,2e-2f,100 trials) and the same collection of real images used in Lian et al. 2017 (2d). |
| Dataset Splits | No | The paper mentions using synthetic and benchmark datasets but does not provide explicit details about training, validation, or test data splits or their percentages/counts. |
| Hardware Specification | Yes | Run on an Intel(R) i7-8650U, 1.9GHz, 16GB machine. |
| Software Dependencies | No | The paper mentions software tools like Macaulay2, CPD, GMMREG, and APM, but it does not specify version numbers for these or for any other software dependencies used in their implementation or experiments. |
| Experiment Setup | Yes | using normally distributed A, x, ε with n = 3, m = 100 and σ = 0.01 for the noise. Algorithm-A stops splitting a hypercube when a depth 6 for that hypercube has been reached. When the affine transformation is a rotation (Fig. 2a) CPD, GMMREG only work for small angles, while they fail for general affine transformations (Figs. 2b-2c). |