Home / Current Issue / Paper 1719374
Development of a Lightweight Temporal Convolutional Network for Microexpression Recognition
Subject area: Science,Engineering and Technology · Area of research: Deep Learning and Computer Vision
DOI: https://doi.org/10.64388/IREV10I1-1719374
Abstract
Microexpression recognition (MER) remains challenging because facial movements are short, subtle, and usually require computationally expensive temporal models. This paper presents a lightweight depthwise-separable Temporal Convolutional Network (DS-TCN) for MER on the CASME II dataset. The model uses a dual-stream design that combines grayscale facial-frame features with optical-flow motion features after face cropping, Eulerian video magnification, resizing, and SSIM-based key-frame selection. Standard temporal convolutions are replaced with depthwise separable temporal convolutions to reduce parameter count and floating-point operations while preserving sequence learning. Three-fold cross-validation produced a validation accuracy of 72.6% and a macro-F1 score of 67.58%. Compared with the standard TCN baseline, the proposed DS-TCN reduced parameters from 1.57 M to 0.219 M, FLOPs from 10.33 G to 1.50 G, and CPU inference time from 580 ms to 210 ms. The results show that lightweight temporal modeling can provide competitive MER performance with substantially lower computational cost.
Keywords
CASME II, Depthwise Separable Convolution, Microexpression Recognition, Optical Flow, Temporal Convolutional Network
How to cite this paper
@article{1719374,
author = {Adeyemi, Ifeoluwa Olamiji, Falohun, Adeleye Samuel, Oguntoye Jonathan Ponmile, Ayinla, Michael Oluwaseun},
title = {Development of a Lightweight Temporal Convolutional Network for Microexpression Recognition},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {1},
pages = {1644-1655},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1719374.pdf},
abstract = {Microexpression recognition (MER) remains challenging because facial movements are short, subtle, and usually require computationally expensive temporal models. This paper presents a lightweight depthwise-separable Temporal Convolutional Network (DS-TCN) for MER on the CASME II dataset. The model uses a dual-stream design that combines grayscale facial-frame features with optical-flow motion features after face cropping, Eulerian video magnification, resizing, and SSIM-based key-frame selection. Standard temporal convolutions are replaced with depthwise separable temporal convolutions to reduce parameter count and floating-point operations while preserving sequence learning. Three-fold cross-validation produced a validation accuracy of 72.6% and a macro-F1 score of 67.58%. Compared with the standard TCN baseline, the proposed DS-TCN reduced parameters from 1.57 M to 0.219 M, FLOPs from 10.33 G to 1.50 G, and CPU inference time from 580 ms to 210 ms. The results show that lightweight temporal modeling can provide competitive MER performance with substantially lower computational cost.},
keywords = {CASME II, Depthwise Separable Convolution, Microexpression Recognition, Optical Flow, Temporal Convolutional Network},
month = {July},
doi = {https://doi.org/10.64388/IREV10I1-1719374}
}