A unique M-pattern for micro-expression spotting in long videos

Authors: Jinxuan Wang, Shiting Xu, Tong Zhang

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on CAS(ME)2, SAMM-LV and CASME II validate the proposed methods are superior to the state-of-the-art methods.
Researcher Affiliation Academia Jinxuan Wang1, Shiting Xu1, Tong Zhang1,2 1South China University of Technology & Engineering Research Center of the Ministry of Education on Health Intelligent Perception and Paralleled Digital-Human 2Pazhou Lab, Guangzhou, China {202221043813,csxst}@mail.scut.edu.cn, tony@scut.edu.cn
Pseudocode Yes Appendix A.1 PSEUDO-CODE FOR FACIAL ALIGNMENT. Algorithm 1 Alignment. Algorithm 2 Reference Frame Update.
Open Source Code No The paper does not provide an explicit statement about the availability of its source code or a link to a code repository.
Open Datasets Yes We conduct the experiments on CASME II(Yan et al., 2014), CAS(ME)2(Qu et al., 2018) and SAMM-LV(Yap et al., 2020).
Dataset Splits No The paper mentions using these datasets for experiments and evaluation, but it does not specify explicit training, validation, or test dataset splits with percentages or counts.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU models, CPU types, memory) used for running the experiments.
Software Dependencies Yes Pixelmatch(Mapbox, 2022) is used for fast pixel-level image comparison. (Reference: Mapbox. pixelmatch 0.3.0. https://pypi.org/project/pixelmatch, 2022. Accessed: 2023-08-01.)
Experiment Setup Yes p is set to 3 for peak detection. For MES, Tmax = {23, 51}, Tddeg = {45, 250}, Io Umerge = {0.3, 0.3}({ , }for CAS(ME)2 and SAMM-LV respectively). Tdisorder = block area / 8 / 8. window length and polyorder is set to k and 3. Mmin and Mmax is set to 0.1% and 8% of the matching box area respectively.