MC-HOG Correlation Tracking with Saliency Proposal
Authors: Guibo Zhu, Jinqiao Wang, Yi Wu, Xiaoyu Zhang, Hanqing Lu
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive evaluations performed on the benchmark dataset show the superiority of the proposed method.1 Introduction Visual tracking, which is to estimate object state in an image sequence, is one of the core problems in computer vision. It has many applications, such as surveillance, action recognition and autonomous robots/car (Yilmaz, Javed, and Shah 2006; Wang et al. 2014). One robust visual tracking approach in real-world scenarios should cope with challenges as much as possible, such as occlusions, background clutter and shape deformation. |
| Researcher Affiliation | Academia | National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China B-DAT & CICAEET, School of Information & Control, Nanjing University of Information Science and Technology, Nanjing, 210044, Jiangsu, China Institute of Information Engineering, Chinese Academy of Sciences, Beijing, 100093, China {gbzhu, jqwang, luhq}@nlpr.ia.ac.cn ywu.china@yahoo.com, zhangxiaoyu@iie.ac.cn |
| Pseudocode | No | The paper presents mathematical equations and descriptions of methods but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available. |
| Open Datasets | Yes | We evaluate our MOCA tracker on the challenging CVPR2013 Visual Tracker Benchmark (Wu, Lim, and Yang 2013), by following rigorously their evaluation protocols. There are totally 50 sequences used to evaluate the proposed approach. |
| Dataset Splits | No | The paper evaluates on the CVPR2013 Visual Tracker Benchmark but does not specify how data within these sequences was split for training, validation, or internal testing. It mentions using 'initial positions in the first frame' for comparison. |
| Hardware Specification | Yes | The experiments are performed in Matlab on an Intel Xeon 2 core 2.50 GHz CPU with 256G RAM. |
| Software Dependencies | No | The paper mentions 'Matlab' as the software environment but does not specify a version number for Matlab or any other software libraries or dependencies used. |
| Experiment Setup | Yes | In all the experiments, we use the same parameter values for all sequences (i.e. λ = 0.0001, γ = 0.1 and βinit = 0.02). |