Kernel Cross-Correlator

Authors: Chen Wang, Le Zhang, Lihua Xie, Junsong Yuan

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Comprehensive experiments on visual tracking and human activity recognition using wearable devices demonstrate its robustness, flexibility, and efficiency.
Researcher Affiliation Collaboration Chen Wang,1 Le Zhang,2 Lihua Xie,1 Junsong Yuan1 1School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 2Advanced Digital Sciences Center, Singapore
Pseudocode No The paper describes methods mathematically but does not include a clearly labeled pseudocode or algorithm block.
Open Source Code Yes The source codes of both experiments are released at https://github.com/wang-chen/KCC.
Open Datasets Yes The well-known benchmark dataset that contains 51 sequences (Wu, Lim, and Yang 2013) is selected. The public wearable action recognition database (WARD) (Yang et al. 2009) is chosen.
Dataset Splits Yes The standard criterion of success plot curve is adopted as the performance criteria. In the experiments, short duration segments (50 instants, about 1.67 s) are randomly selected from the database for each trail. Then the cross-validation with single training sample (one segment) is performed among all 13 activities.
Hardware Specification No No specific hardware details (like GPU/CPU models, memory, or processor types) used for running the experiments are mentioned in the paper.
Software Dependencies No No specific software dependencies with version numbers are mentioned in the paper.
Experiment Setup Yes In both correlators, the regularization term λ in (2) is set as 0.001 and the Gaussian kernel is used with sigma of 0.1. In the experiments, the regularization term λ in (2) is set as 0.0015 while the Gaussian kernel is used with sigma of 1.