Hidden Follower Detection: How Is the Gaze-Spacing Pattern Embodied in Frequency Domain?

Authors: Shu Li, Ruimin Hu, Suhui Li, Liang Liao

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate the feature fidelity of time series and HFD performance are positively correlated, and the fidelity of frequencydomain features and HFD performance are significantly better than the time-domain features. On both real and simulated datasets, the accuracy of the proposed method is increased by 3%, and the gaze-only module is improved by 10%.
Researcher Affiliation Academia 1School of Cyber Engineering, Xidian University 2School of Computer Science and Engineering, Nanyang Technological University
Pseudocode No The paper describes methods and calculations but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access information (e.g., repository link, explicit statement of code release, or mention of code in supplementary materials) for the methodology described.
Open Datasets Yes Real-HFD. ... More details can be found in (Xu et al. 2021, 2022).
Dataset Splits Yes We trained with 4-fold cross-validation, of which the ratio of the training set and test set in both datasets is 7:3.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running its experiments.
Software Dependencies No The paper mentions software tools and models like Tra De S, Dense Pose, Gaze360, and TWIESN, but does not specify their version numbers or other software dependencies with specific versions.
Experiment Setup Yes The optimal gaze and spacing MFCC parameters in Fig. 1 are: gaze MFCCs with dimension = 4, w = 10s, overlap = 2s and spacing MFCCs with dimension = 4, w = 8s, overlap = 1s. ... If P>0.6, the pedestrian is recognized as a hidden follower. ... For the time-domain features, we try to reconstruct using an autoregressive model: Transformer, we keep the network structure, set b L as the outputting reconstructed series, and introduce the SDR as the loss function.