Path Following with Adaptive Path Estimation for Graph Matching
Authors: Tao Wang, Haibin Ling
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To validate our approach, we compare our approach with several recently proposed graph matching algorithms on three benchmark image datasets. Experimental results show that, our approach improves significantly the computation efficiency of the original algorithms, and offers similar or better matching results. |
| Researcher Affiliation | Collaboration | Tao Wang1 Haibin Ling2,3 1 School of Computer &Information Technology, Beijing Jiaotong University, Beijing 100044, China 2 Computer & Information Sciences Department, Temple University, Philadelphia 19122, USA 3 Meitu Hi Scene Lab, Hi Scene Information Technologies, Shanghai, China |
| Pseudocode | Yes | Algorithm 1: Path Estimation(x, d, t, k ) and Algorithm 2: GNCCP+EST, updated GNCCP algorithm are present in the paper. |
| Open Source Code | No | The paper does not contain any explicit statements about making the source code for the described methodology publicly available, nor does it provide links to a code repository. |
| Open Datasets | Yes | We compare the proposed FGM+EST and GNCCP+EST algorithms with the original FGM and GNCCP algorithms and three recent graph matching algorithms, IPFP (Leordeanu and Hebert 2009), RRWM (Cho, Lee, and Lee 2010) and PSM (Egozi, Keller, and Guterman 2013), and report experimental results on three benchmark datasets. The CMU house image sequences is commonly used to test the performance of graph matching algorithms (Cho, Lee, and Lee 2010; Zhou and Torre 2012; Duchenne, Joulin, and Ponce 2011; Torresani, Kolmogorov, and Rother 2008). This dataset from (Leordeanu, Sukthankar, and Hebert 2012) consists of 30 pairs of car images and 20 pairs of motorbike images selected from Pascal 2007. In this experiment, we test our approach on a real image dataset containing 30 image pairs provided in (Cho, Lee, and Lee 2010) which are collected from Caltech-101 and MSRC datasets. |
| Dataset Splits | No | The paper describes how test scenarios are set up (e.g., varying outliers, sequence gaps) but does not provide specific details on train/validation/test splits (e.g., percentages, exact sample counts) or cross-validation methods that would be needed for reproduction. |
| Hardware Specification | No | The paper does not specify any details about the hardware (e.g., GPU models, CPU types, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific version numbers for any software dependencies or libraries used in the implementation of the proposed methods or experiments. |
| Experiment Setup | Yes | The parameter k controls the number of previous solution points that are used to estimate the next solution point. We set k = 10, dmin = 0.002 and θ = 0.001 throughout our experiments. |