Nonparametric Continuous Sensor Registration
Authors: William Clark, Maani Ghaffari, Anthony Bloch
JMLR 2021 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Low-dimensional cases are derived with numerical examples to show the generality of the proposed framework; A specific derivation and implementation of this framework for RGB-D cameras performs well in texture and structure-scarce environments. Section 9. Experimental Results: RGB-D Visual Odometry includes tables (Table 6, Table 7) showing RMSE values, indicating empirical evaluation. |
| Researcher Affiliation | Academia | William Clark EMAIL Department of Mathematics Cornell University Ithaca, NY 14850, USA Maani Ghaffari EMAIL Department of Naval Architecture and Marine Engineering University of Michigan Ann Arbor, MI 48109 USA Anthony Bloch EMAIL Department of Mathematics University of Michigan Ann Arbor, MI 48109 USA |
| Pseudocode | No | The paper describes mathematical derivations and methodologies in detail using prose and equations, but it does not contain a clearly labeled 'Pseudocode' or 'Algorithm' block, nor does it present structured steps formatted like code. |
| Open Source Code | Yes | 4. the open-source implementation of the derived cases in this work is available for download at https://github.com/Maani Ghaffari/c-sensor-registration https://github.com/Maani Ghaffari/cvo-rgbd |
| Open Datasets | Yes | We performed experiments on two parts of RGB-D SLAM data set and benchmark by the Technical University of Munich (Sturm et al., 2012). This data set was collected indoors with a Microsoft Kinect using a motion capture system as a proxy for ground truth trajectory. |
| Dataset Splits | Yes | We evaluated CVO and DVO on the training and validation sets for all the fr1 sequences and the structure versus texture sequences. |
| Hardware Specification | Yes | All experiments are performed on a Dell XPS15 9750 laptop with Intel i7-8750H CPU (6 cores with 2.20 GHz each) and 32GB RAM. |
| Software Dependencies | Yes | The source code is implemented in C++ and compiled with the Intel Compiler. The kernel computations are parallelized using the Intel Threading Building Blocks (TBB) (Intel Corporation, 2019). |
| Experiment Setup | Yes | Table 5: Parameters used for evaluation using TUM RGB-D Benchmark, similar values are chosen for all experiments. The kernel characteristic length-scale is chosen to be adaptive as the algorithm converges (Ghaffari et al., 2019); intuitively, we prefer a large neighborhood of correlation for each point, but as the algorithm reaches the convergence reducing the local correlation neighborhood allows for faster convergence and better refinement. |