Pith. sign in

REVIEW 2 major objections 75 references

STAG: Spatio-temporal Evolving Structural Representation of Action Units for Micro-expression Recognition

T0 review · 2 major / 0 minor · reviewed 2026-06-29 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read STAG couples AU-guided dynamic graph connectivity with bidirectional cross-attention to capture subtle facial dynamics for micro-expression recognition across datasets.

desk verdict STAG combines graph attention with transformers and AU-guided dynamic edges for micro-expressions, but the abstract supplies zero numbers so the performance claims cannot be checked. read the letter →

arxiv 2606.28083 v1 pith:PSQ4MFXV submitted 2026-06-26 cs.CV cs.AIcs.GRcs.HCcs.MM

Nandani Sharma , Varun Sharma , Dinesh Singh This is my paper · ORCID
classification cs.CVcs.AIcs.GRcs.HCcs.MM
keywords micro-expressionrecognitionactionunitsspatio-temporalnetworkgraphattentiontransformerencoderbidirectionalcross-attentionopticalflowdynamicconnectivity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper aims to show that micro-expression recognition improves when a network jointly models evolving spatial connections between facial regions and their temporal flow instead of treating space and time separately or fixing on apex frames. It selects frames by motion magnitude, builds an action-unit-guided graph whose edges adapt to muscle activations, runs a graph attention branch for structure and a transformer for sequence, then lets the two branches refine each other through bidirectional cross-attention before fusing the result. A sympathetic reader would care because current systems often miss the brief, low-intensity movements that reveal genuine emotion, and a method that generalizes better across recording conditions could support more reliable use in psychology, security, and interaction studies. If the central mechanisms work as described, the fused representation should deliver measurable gains in accuracy, robustness, and interpretability on standard benchmarks.

What carries the argument

AU-guided dynamic connectivity, which adapts the interactions among facial regions inside an enhanced graph attention network according to muscle activation patterns while bidirectional cross-attention refines the resulting spatial features against temporal transformer outputs.

What would settle it

A new, previously unseen dataset on which STAG shows no statistically significant accuracy gain over standard spatio-temporal baselines or exhibits clear signs of overfitting to the training distributions would falsify the generalization claim.

Watch

Extended reading notes

Core claim

STAG is a dynamic ROI-AU-coupled spatial-temporal network that extracts optical flow from magnitude-selected frames, employs an enhanced graph attention network whose connectivity is guided by action-unit activation patterns for spatial reasoning, uses a transformer encoder for temporal modeling, and applies bidirectional cross-attention to mutually refine the spatial and temporal features before final fusion with focal loss. The framework is evaluated on CASME II, 4DME, DFME, NaME, SAMM, and SMIC-HS and is claimed to produce improved robustness, generalization, interpretability, and efficiency through adaptive relational reasoning and deep spatial-temporal fusion.

Load-bearing premise

That magnitude-based frame selection together with AU-guided dynamic connectivity and bidirectional cross-attention will capture the necessary subtle inter-frame dynamics and generalize without dataset-specific overfitting or hidden biases.

Editorial extensions

If this is right

  • Magnitude-based frame selection plus transformer temporal modeling captures inter-frame dynamics that apex-frame methods miss.
  • AU-guided dynamic connectivity produces more interpretable spatial reasoning tied directly to muscle activations.
  • Bidirectional cross-attention yields fused representations that improve cross-dataset performance on CASME II, SAMM, and similar collections.
  • The overall pipeline reports gains in both recognition accuracy and computational efficiency.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same adaptive graph-plus-attention pattern could be tested on other subtle motion domains such as micro-gestures or medical movement analysis.
  • Replacing the fixed AU guidance with learned activation predictors might reduce reliance on pre-annotated action units.
  • The bidirectional refinement module offers a template for mutual spatial-temporal learning that could be applied to video tasks outside facial analysis.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 0 minor

Summary. The paper proposes STAG, a dynamic ROI-AU-coupled spatial-temporal network for micro-expression recognition. It extracts optical flow from magnitude-selected frames, employs a dual-branch architecture (enhanced GAT for spatial reasoning via AU-guided dynamic connectivity and transformer for temporal modeling), uses bidirectional cross-attention for mutual feature refinement, and optimizes with focal loss. The method is claimed to improve robustness, generalization, interpretability, and efficiency, with evaluation on CASME II, 4DME, DFME, NaME, SAMM, and SMIC-HS demonstrating the effectiveness of adaptive relational reasoning and deep spatial-temporal fusion for cross-dataset recognition.

Significance. If substantiated, the work could advance micro-expression recognition by addressing limitations of apex-frame reliance and separate spatial-temporal modeling through adaptive AU-guided graphs and cross-attention fusion. This might yield better handling of subtle dynamics and cross-dataset performance in a field where generalization remains challenging.

major comments (2)
  1. [Abstract] Abstract: The abstract asserts that 'extensive experiments demonstrate improved robustness, generalization, interpretability, and computational efficiency' across six datasets and confirms effectiveness of the proposed components, but supplies no quantitative results, baseline comparisons, error bars, or statistical tests. This prevents verification of the central empirical claim.
  2. [Abstract] The central assumption that magnitude-based frame selection, AU-guided dynamic connectivity, and bidirectional cross-attention will capture subtle inter-frame dynamics and generalize without dataset-specific overfitting or unaccounted biases is load-bearing for the cross-dataset claims, yet no details on validation strategy, ablation studies, or bias controls are referenced to support it.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed feedback on the abstract. We agree that the abstract would benefit from greater specificity to support the empirical claims and will revise it accordingly in the next version of the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The abstract asserts that 'extensive experiments demonstrate improved robustness, generalization, interpretability, and computational efficiency' across six datasets and confirms effectiveness of the proposed components, but supplies no quantitative results, baseline comparisons, error bars, or statistical tests. This prevents verification of the central empirical claim.

    Authors: We acknowledge the limitation in the current abstract. The full manuscript contains quantitative results in Section 5, including per-dataset accuracies, comparisons against multiple baselines, ablation tables, and statistical significance tests. We will revise the abstract to incorporate key quantitative highlights (e.g., accuracy gains and cross-dataset averages) while respecting length constraints. revision: yes

  2. Referee: [Abstract] The central assumption that magnitude-based frame selection, AU-guided dynamic connectivity, and bidirectional cross-attention will capture subtle inter-frame dynamics and generalize without dataset-specific overfitting or unaccounted biases is load-bearing for the cross-dataset claims, yet no details on validation strategy, ablation studies, or bias controls are referenced to support it.

    Authors: The validation protocol (including cross-dataset splits and leave-one-subject-out), ablation studies on each component, and bias considerations are presented in Sections 4 and 5. We will update the abstract to briefly reference the cross-dataset evaluation strategy and component ablations that underpin the generalization claims. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity

full rationale

The paper is an empirical architecture proposal for micro-expression recognition. It describes a network with optical flow, graph attention, transformers, bidirectional cross-attention, and AU-guided connectivity, evaluated via experiments on standard datasets. No equations, derivations, fitted parameters presented as predictions, or self-citation chains appear in the provided text. The central claims rest on cross-dataset performance results rather than any reduction of outputs to inputs by construction. This matches the most common honest finding of a self-contained empirical method.

Assumptions & free parameters 0 free parameters · 0 assumptions · 0 invented entities

Abstract contains no mathematical derivations, free parameters, axioms, or invented entities; ledger remains empty due to lack of technical detail.

how reviews work

0 comments
Cite this review

Pith. "Pith review of STAG: Spatio-temporal Evolving Structural Representation of Action Units for Micro-expression Recognition." pith.science (2026). https://pith.science/paper/PSQ4MFXV

@misc{pith2026260628083,
  author       = {Pith},
  title        = {Pith review of: STAG: Spatio-temporal Evolving Structural Representation of Action Units for Micro-expression Recognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PSQ4MFXV}},
  note         = {Machine review of arXiv:2606.28083}
}
read the original abstract

Micro-expression recognition is challenging due to subtle and short-lived facial muscle movements. Existing methods rely heavily on apex-onset frames, overlook fine-grained inter-frame dynamics, and separately model spatial and temporal information, limiting generalization across datasets. To address these challenges, we propose STAG, a dynamic ROI-AU-coupled spatial-temporal network that jointly models motion flow and adaptive facial connectivity. The framework extracts optical flow from discriminative frames using magnitude-based selection and temporal attention. A dual-branch architecture combines an enhanced graph attention network for structured spatial reasoning with a transformer encoder for temporal modeling. A bidirectional cross-attention module enables mutual refinement of spatial and temporal features, while AU-guided dynamic connectivity adapts facial region interactions according to muscle activation patterns. The transformer captures subtle temporal dynamics beyond apex-based approaches, improving semantic consistency and interpretability for explainable micro-expression recognition. The fused representation is optimized using focal loss and evaluated on CASME II, 4DME, DFME, NaME, SAMM, and SMIC-HS. Extensive experiments demonstrate improved robustness, generalization, interpretability, and computational efficiency, confirming the effectiveness of adaptive relational reasoning, AU-guided dynamic connectivity, and deep spatial-temporal feature fusion for accurate cross-dataset micro-expression recognition.

Figures

Figures reproduced from arXiv: 2606.28083 by the authors.

Figure 1
Figure 1. Overview of the proposed STAG framework for MER. Given an input facial video sequence, facial landmarks are detected and ROIs are constructed around key facial components, including the eyes, eyebrows, and mouth. Dense optical flow is then computed to capture subtle facial motions, producing motion intensity maps and flow vectors represented as feature tensor 𝐗. These features are projected into a latent embedding s… view at source ↗
Figure 2
Figure 2. Explainability analysis of the proposed STAG framework across multiple datasets using Grad-CAM, perturbation maps, optical flow, and E-GAT attention visualizations. Results are shown for positive classes (SMIC-HS and SAMM) and negative classes (remaining datasets), highlighting the model’s focus on discriminative facial action regions and motion patterns [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
Figure 3
Figure 3. Qualitative comparison of heat-maps generated by different SOTA MER methods for the [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Visualization of ROI-based optical flow analysis for a micro-expression sample predicted as mentioned in the [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: t-SNE visualization of learned feature embeddings. The proposed STAG method produces well-separated clusters for different emotion classes, indicating strong discriminative capability. Compared to baseline representations, STAG reduces class overlap and improves intra-…
Figure 6
Figure 6. Figure 6: Spatial connection and message-passing visualization across the two sequential layers of the E-GAT. Node positions correspond structurally to facial ROIs, where edge colors and thicknesses signify the relative attention weights dynamically assigned to spatial dependenc…
Figure 7
Figure 7. Figure 7: Performance comparison of the proposed STAG framework across multiple MER benchmark datasets under different evaluation protocols (LOSO, SKF, and GKF). The plot illustrates the variation of UF1, UAR, and WAR scores with respect to dataset complexity and class settings …
Figure 8
Figure 8. Figure 8: Confusion matrices of STAG on six benchmark micro-expression datasets (4DME, CASME-II, DFME, SAMM, SMIC-HS, and NaME) for 3-class (top) and 5-class (bottom) classification tasks. Strong diagonal patterns indicate high recognition accuracy, while off-diagonal entries hi…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

75 extracted references · 3 canonical work pages

  1. [1]

    Boosting micro-expression recognition via self-expression reconstruction and memory contrastive learning

    Bao,Y.,Wu,C.,Zhang,P.,Shan,C.,Qi,Y.,Ben,X.,2024. Boosting micro-expression recognition via self-expression reconstruction and memory contrastive learning. IEEE Transactions on Affective Com- puting 15, 2083–2096

    2024

  2. [2]

    SAMM:Aspontaneousmicro-facialmovementdataset

    Davison, A.K., Lansley, C., Costen, N., Tan, K., Yap, M.H., 2018. SAMM:Aspontaneousmicro-facialmovementdataset. IEEETrans- actions on Affective Computing 9, 116–129

    2018

  3. [3]

    Facial action coding system

    Ekman, P., Friesen, W.V., 1978. Facial action coding system. Con- sulting Psychologists Press

    1978

  4. [4]

    Graph-driven micro-expression rendering with emotionally diverse expressions for lifelike digital humans

    Fang,L.,Yang,F.,Lin,Y.,Zhang,J.,Whang,M.,2025. Graph-driven micro-expression rendering with emotionally diverse expressions for lifelike digital humans. Biomimetics 10, 587

    2025

  5. [5]

    Off- apexnet on micro-expression recognition system

    Gan,Y.S.,Liong,S.T.,Yau,W.C.,Huang,Y.C.,Tan,L.K.,2019. Off- apexnet on micro-expression recognition system. Signal Processing: Image Communication 74, 129–139

    2019

  6. [6]

    Irawan, B., Utama, N.P., Munir, R., Purwarianti, A., 2023. Sponta- neous micro-expression recognition using 3dcnn on long videos for emotionanalysis,in:202310thinternationalconferenceonadvanced informatics: concept, theory and application (ICAICTA), IEEE. pp. 1–6. N. Sharma et al.:Preprint submitted to ElsevierPage 18 of 21 Spatio-temporal Evolving Struct...

    2023

  7. [7]

    Highlyeffectiveend-to-endsingle-to-multichannelfeaturefusionand ensemble classification to decode emotional secretes from small- scale spontaneous facial micro-expressions

    Islam, M.S., Sang, Y., Mohammed, A.A., Yuan, L., Lv, J., 2023. Highlyeffectiveend-to-endsingle-to-multichannelfeaturefusionand ensemble classification to decode emotional secretes from small- scale spontaneous facial micro-expressions. Journal of King Saud University-Computer and Information Sciences 35, 101653

    2023

  8. [8]

    Izmailov, P., Podoprikhin, D., Garipov, T., Vetrov, D., Wilson, A.G.,

Show all 75 references
  1. [9]

    Averagingweightsleadstowideroptimaandbettergeneraliza- tion, in: UAI

  2. [10]

    Dlib-ml: A machine learning toolkit

    King, D.E., 2009. Dlib-ml: A machine learning toolkit. Journal of Machine Learning Research 10, 1755–1758

    2009

  3. [11]

    Semi-supervised classification with graphconvolutionalnetworks,in:InternationalConferenceonLearn- ing Representations (ICLR)

    Kipf, T.N., Welling, M., 2017. Semi-supervised classification with graphconvolutionalnetworks,in:InternationalConferenceonLearn- ing Representations (ICLR)

    2017

  4. [12]

    3d micro-expression recognition based on adaptive dynamic vision

    Kong, W., You, Z., Lv, X., 2025. 3d micro-expression recognition based on adaptive dynamic vision. Sensors 25, 3175

    2025

  5. [13]

    Temporaldiscrimina- tive micro-expression recognition via graph contrastive learning, in: 26th International Conference on Pattern Recognition (ICPR), IEEE

    Lao,L.,Li,Y.,Liu,M.L.,Xu,C.,Cui,Z.,2022. Temporaldiscrimina- tive micro-expression recognition via graph contrastive learning, in: 26th International Conference on Pattern Recognition (ICPR), IEEE. pp. 1033–1040. N. Sharma et al.:Preprint submitted to ElsevierPage 19 of 21 Spa...

    2022

  6. [14]

    Lei, L., Chen, T., Li, S., Li, J., 2021. Micro-expression recognition based on facial graph representation learning and facial action unit fusion, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 1571–1580

    2021

  7. [15]

    Anovelgraphneuralnetworkmethodformicro- expression recognition

    Lei,L.,etal.,2020. Anovelgraphneuralnetworkmethodformicro- expression recognition. Information Sciences

    2020

  8. [16]

    Structure representation with adaptive and compact facial graph for micro- expression recognition

    Li,C., Ba,R.,Wang, X.,Yu,M., Li,X.,Huang, D.,2025a. Structure representation with adaptive and compact facial graph for micro- expression recognition. IEEE Transactions on Biometrics, Behavior, and Identity Science 7, 256–269

  9. [17]

    Li, J., Qian, Y., Zhao, L., Wang, S.J., 2025b. Fed-psyau: Privacy- preserving micro-expression recognition via psychological au coor- dination and dynamic facial motion modeling, in: Proceedings of the IEEE/CVFInternationalConferenceonComputerVision,pp.14453– 14463

  10. [18]

    Micro-expression recognition through feature enhancement and region weighted fusion based on supervised contrastive learning

    Li, S., Li, M., Sun, J., Lu, S., 2026. Micro-expression recognition through feature enhancement and region weighted fusion based on supervised contrastive learning. Signal Processing 238, 110171

    2026

  11. [19]

    4dme: A spontaneous 4d micro-expression dataset with multimodalities

    Li, X., Cheng, S., Li, Y., Behzad, M., Shen, J., Zafeiriou, S., Pantic, M., Zhao, G., 2022a. 4dme: A spontaneous 4d micro-expression dataset with multimodalities. IEEE Transactions on Affective Com- puting 14, 3031–3047

  12. [20]

    A spontaneous micro-expression database: Inducement, collection and baseline, in: Face and Gesture 2013, IEEE

    Li, X., Pfister, T., Huang, X., Zhao, G., Pietikäinen, M., 2013. A spontaneous micro-expression database: Inducement, collection and baseline, in: Face and Gesture 2013, IEEE. pp. 1–6

    2013

  13. [21]

    Micro-expression recognition: A survey

    Li, X., et al., 2022b. Micro-expression recognition: A survey. IEEE Transactions on Affective Computing

  14. [22]

    Liang, H., Liu, J., Wang, Y., Shi, H., 2025. Cit: Context interaction transformer for micro-expression recognition in natural scene, in: 20254thInternationalConferenceonArtificialIntelligence,Human- Computer Interaction and Robotics (AIHCIR), IEEE. pp. 1–6

    2025

  15. [23]

    Journal of Signal Processing Systems 92, 705–725

    Liong, S.T., Gan, Y.S., Zheng, D., Li, S.M., Xu, H.X., Zhang, H.Z., Lyu,R.K.,Liu,K.H.,2020.Evaluationofthespatio-temporalfeatures and gan for micro-expression recognition system. Journal of Signal Processing Systems 92, 705–725

    2020

  16. [24]

    Temporal-spatial cross- fusionfordynamicmicroexpressionrecognition.PatternRecognition , 113715

    Liu, F., Nan, B., Qian, X., Fu, X., 2026. Temporal-spatial cross- fusionfordynamicmicroexpressionrecognition.PatternRecognition , 113715

    2026

  17. [25]

    Name: A natural micro-expression dataset for micro- expression recognition in the wild, in: Proceedings of the 33rd ACM International Conference on Multimedia, pp

    Liu, J., Shi, H., Liang, H., Xu, X., Zong, Y., Wang, Y., Zheng, W., 2025a. Name: A natural micro-expression dataset for micro- expression recognition in the wild, in: Proceedings of the 33rd ACM International Conference on Multimedia, pp. 5617–5626

  18. [26]

    Mer-clip: Au-guided vision-language alignment for micro-expression recogni- tion

    Liu, S., Mao, X., Zhao, S., Li, P., Xu, T., Chen, E., 2025b. Mer-clip: Au-guided vision-language alignment for micro-expression recogni- tion. IEEE Transactions on Affective Computing

  19. [27]

    Decoupled weight decay regulariza- tion, in: ICLR

    Loshchilov, I., Hutter, F., 2019. Decoupled weight decay regulariza- tion, in: ICLR

    2019

  20. [28]

    Ma, C., Zhao, S., Pei, Y., Xie, L., Yin, E., Yan, Y., 2025. A multi- prior fusion network for video-based micro-expression recognition, in: ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE. pp. 1–5

    2025

  21. [29]

    Acomprehensivereview of microexpression recognition, classification, and datasets

    Malik,P.,Singh,J.,Ali,F.,Kwak,D.,2026. Acomprehensivereview of microexpression recognition, classification, and datasets. IEEE Transactions on Computational Social Systems

    2026

  22. [30]

    Time efficient micro-expression recognition using weighted spatio- temporal landmark graphs, in: International Conference on Machine Learning and Applications (ICMLA), IEEE

    Matharaarachchi, N., Pasha, M.F., Coleman, S., Kerr, D., 2023. Time efficient micro-expression recognition using weighted spatio- temporal landmark graphs, in: International Conference on Machine Learning and Applications (ICMLA), IEEE. pp. 347–354

    2023

  23. [31]

    Shangguan, Z., Dong, Y., Guo, S., Leung, V.C., Deen, M.J., Hu, X.,

  24. [32]

    ACM Computing Surveys 58, 1–39

    Facial expression analysis and its potentials in iot systems: A contemporary survey. ACM Computing Surveys 58, 1–39

  25. [33]

    Mol:Jointestimationofmicro-expression,opticalflow,andlandmark via transformer-graph-style convolution

    Shao, Z., Cheng, Y., Li, F., Zhou, Y., Lu, X., Xie, Y., Ma, L., 2025a. Mol:Jointestimationofmicro-expression,opticalflow,andlandmark via transformer-graph-style convolution. arXiv preprint

  26. [34]

    Micro-expression recognition via fine-grained dynamic perception

    Shao, Z., Cheng, Y., Zhang, F., Shi, X., Li, C., Ma, L., Yeung, D.Y., 2025b. Micro-expression recognition via fine-grained dynamic perception. ACM Transactions on Multimedia Computing, Commu- nications and Applications 21, 1–23

  27. [35]

    High-level lora and hierarchical fusion for enhanced micro- expression recognition

    Shao,Z.,Cheng,Y.,Zhou,Y.,Xiang,X.,Li,J.,Liu,B.,Yeung,D.Y., 2025c. High-level lora and hierarchical fusion for enhanced micro- expression recognition. The Visual Computer 41, 4533–4546

  28. [36]

    Improvedsingle- stage facial landmarks for real-time driver drowsiness detection, in: International Conference on Distributed Computing and Intelligent Technology, Springer

    Sharma,N.,Banothu,S.,Shah,M.,Singh,D.,2026. Improvedsingle- stage facial landmarks for real-time driver drowsiness detection, in: International Conference on Distributed Computing and Intelligent Technology, Springer. pp. 189–204

    2026

  29. [37]

    Exp-graph: How connections learn facialattributesingraph-basedexpressionrecognition

    Sharma, N., Singh, D., 2025. Exp-graph: How connections learn facialattributesingraph-basedexpressionrecognition. arXivpreprint arXiv:2507.14608

    2025

  30. [38]

    Para-x: Graph-based facial paralysisdetectionusingstructuraldeformationsoffacialexpression, in:2025InternationalJointConferenceonNeuralNetworks(IJCNN), IEEE

    Sharma, N., Singh, K., Singh, D., 2025. Para-x: Graph-based facial paralysisdetectionusingstructuraldeformationsoffacialexpression, in:2025InternationalJointConferenceonNeuralNetworks(IJCNN), IEEE. pp. 1–8

    2025

  31. [39]

    Dsbic- net:dynamic-staticbidirectionalinteractioncollaborativenetworkfor micro-expression recognition

    Shou, Z., Huang, L., Yuan, X., Yu, Y., Xu, X., Wu, Z., 2026. Dsbic- net:dynamic-staticbidirectionalinteractioncollaborativenetworkfor micro-expression recognition. Multimedia Systems 32, 122

    2026

  32. [40]

    Micro-expressionrecogni- tion using a shallow convlstm-based network, in: Proceedings of the Asian Conference on Computer Vision, pp

    Shukla,S.,Rai,P.K.,Verlekar,T.T.,2022. Micro-expressionrecogni- tion using a shallow convlstm-based network, in: Proceedings of the Asian Conference on Computer Vision, pp. 17–28

    2022

  33. [41]

    Facialmicro-expressionrecognitionusing stochasticgraphconvolutionalnetworkanddualtransferredlearning

    Tang,H.,Chai,L.,2024a. Facialmicro-expressionrecognitionusing stochasticgraphconvolutionalnetworkanddualtransferredlearning. Neural Networks 178, 106421

  34. [42]

    Facialmicro-expressionrecognitionusing stochasticgraphconvolutionalnetworkanddualtransferredlearning

    Tang,H.,Chai,L.,2024b. Facialmicro-expressionrecognitionusing stochasticgraphconvolutionalnetworkanddualtransferredlearning. Neural Networks 178, 106421. URL:https://doi.org/10.1016/j. neunet.2024.106421, doi:10.1016/J.NEUNET.2024.106421

    2024 doi

  35. [43]

    Attention is all you need

    Vaswani,A.,Shazeer,N.,Parmar,N.,Uszkoreit,J.,Jones,L.,Gomez, A.N., Kaiser, Ł., Polosukhin, I., 2017. Attention is all you need. Advances in Neural Information Processing Systems 30

    2017

  36. [44]

    Me-nas: A micro expression feature adaptive neural architecture search

    Verma, M., Vipparthi, S.K., Murala, S., Abdel-Mottaleb, M., 2026. Me-nas: A micro expression feature adaptive neural architecture search. ACM Transactions on Intelligent Systems and Technology

    2026

  37. [45]

    A multi-modal multi-scalenetworkbasedontransformerformicro-expressionrecog- nition

    Wang, F., Li, J., Qi, C., Wang, L., Wang, P., 2025a. A multi-modal multi-scalenetworkbasedontransformerformicro-expressionrecog- nition. JournalofVisualCommunicationandImageRepresentation, 104537

  38. [46]

    Dbde-net: Dual-branch detail-enhanced network for micro-expression recognition

    Wang, H., Wang, L., Xu, L., Li, Y., 2026a. Dbde-net: Dual-branch detail-enhanced network for micro-expression recognition. Neuro- computing , 133167

  39. [47]

    Gta: A graph- transformer architecture for micro-expression recognition, in: IEEE International Conference on Unmanned Systems (ICUS), pp

    Wang, K., Xu, H., Wu, M., Yao, X., Zhang, P., 2025b. Gta: A graph- transformer architecture for micro-expression recognition, in: IEEE International Conference on Unmanned Systems (ICUS), pp. 1611– 1616

  40. [48]

    Micro-expression recognitionvialora-enhanceddinov2andinteractivespatio-temporal modeling

    Wang, M., Tang, X., Wang, B., Ren, J., 2026b. Micro-expression recognitionvialora-enhanceddinov2andinteractivespatio-temporal modeling. Sensors 26, 625

  41. [49]

    Wang,Y.,Huang,Y.,Liu,C.,Gu,X.,Yang,D.,Wang,S.,Zhang,B.,

  42. [50]

    JournalofHealthcareEngineering2021,7799100

    Micro expression recognition via dual-stream spatiotemporal attentionnetwork. JournalofHealthcareEngineering2021,7799100

  43. [51]

    Wang, Z., Yang, M., Jiao, Q., Xu, L., Han, B., Li, Y., Tan, X.,

  44. [52]

    Sensors 24, 1574

    Two-level spatio-temporal feature fused two-stream network for micro-expression recognition. Sensors 24, 1574

  45. [53]

    Geometric graph representation with learnable graph structure and adaptive au constraint for micro-expression recognition

    Wei,J.,Peng,W.,Lu,G.,Li,Y.,Yan,J.,Zhao,G.,2024a. Geometric graph representation with learnable graph structure and adaptive au constraint for micro-expression recognition. IEEE Transactions on Affective Computing 15, 1343–1357

  46. [54]

    Geometric graph representation with learnable graph structure and adaptive AU constraint for micro-expression recognition

    Wei,J.,Peng,W.,Lu,G.,Li,Y.,Yan,J.,Zhao,G.,2024b. Geometric graph representation with learnable graph structure and adaptive AU constraint for micro-expression recognition. IEEE Trans. Affect. Comput.15,1343–1357. URL:https://doi.org/10.1109/TAFFC.2023. 3340016, doi:10.1109/TAF...

    2023 doi

  47. [55]

    Wei, M., Zheng, W., Zong, Y., Jiang, X., Lu, C., Liu, J., 2022. A novel micro-expression recognition approach using attention-based magnification-adaptive networks, in: ICASSP 2022-2022 IEEE In- ternational Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE. ...

    2022

  48. [56]

    Revealing the invisible with model and data shrinking for composite-database micro-expression recognition

    Xia, Z., Peng, W., Khor, H.Q., Feng, X., Zhao, G., 2020. Revealing the invisible with model and data shrinking for composite-database micro-expression recognition. IEEE Transactions on Image Process- ing 29, 8590–8605

    2020

  49. [57]

    Assisted macroscopic expression recognition based on graphs

    Xie, X., et al., 2020. Assisted macroscopic expression recognition based on graphs. Pattern Recognition Letters

    2020

  50. [58]

    Motion-aware representation network for micro-expression recognition

    Xu, S., Chen, C.P., Zhang, T., 2026. Motion-aware representation network for micro-expression recognition. IEEE Transactions on Multimedia

    2026

  51. [59]

    CASMEII:Andatabaseforspontaneousmacro-expression andmicro-expressionspottingandrecognition

    Yan, W.J., Li, X., Wang, S.J., Zhao, G., Liu, Y.J., Chen, Y.H., Hao, X.,2014. CASMEII:Andatabaseforspontaneousmacro-expression andmicro-expressionspottingandrecognition. IEEETransactionson Affective Computing 5, 3–14

    2014

  52. [60]

    Deep learning pipeline for spotting macro-and micro-expressions in long video sequences based on action units and optical flow

    Yang, B., Wu, J., Ikeda, K., Hattori, G., Sugano, M., Iwasawa, Y., Matsuo, Y., 2023. Deep learning pipeline for spotting macro-and micro-expressions in long video sequences based on action units and optical flow. Pattern Recognition Letters 165, 63–74

    2023

  53. [61]

    Sofp: Capturing subtle facial dynamics with symmetric optical flow perception for micro- expression recognition

    Yu, K., Zhang, Z., Hu, C., Luo, J., 2026. Sofp: Capturing subtle facial dynamics with symmetric optical flow perception for micro- expression recognition. Pattern Recognition , 113199

    2026

  54. [62]

    Lssnet: A two-stream convolu- tionalneuralnetworkforspottingmacro-andmicro-expressioninlong videos,in:Proceedingsofthe29thACMInternationalConferenceon Multimedia, pp

    Yu, W.W., Jiang, J., Li, Y.J., 2021. Lssnet: A two-stream convolu- tionalneuralnetworkforspottingmacro-andmicro-expressioninlong videos,in:Proceedingsofthe29thACMInternationalConferenceon Multimedia, pp. 4745–4749

    2021

  55. [63]

    Dynamic stereotype theory induced micro-expression recognition with oriented deforma- tion,in:ProceedingsoftheComputerVisionandPatternRecognition Conference, pp

    Zhang, B., Wang, X., Wang, C., He, G., 2025a. Dynamic stereotype theory induced micro-expression recognition with oriented deforma- tion,in:ProceedingsoftheComputerVisionandPatternRecognition Conference, pp. 10701–10711

  56. [64]

    Adaptive tempo- ral motion guided graph convolution network for micro-expression recognition, in: IEEE International Conference on Multimedia and Expo (ICME), pp

    Zhang, F., Huang, Z., Zhang, X., Jin, Q., 2024. Adaptive tempo- ral motion guided graph convolution network for micro-expression recognition, in: IEEE International Conference on Multimedia and Expo (ICME), pp. 1–6

    2024

  57. [65]

    Micro-expressionrecognitionbasedondirectlearningofgraphstruc- ture

    Zhang, L., Zhang, Y., Sun, X., Tang, W., Wang, X., Li, Z., 2025b. Micro-expressionrecognitionbasedondirectlearningofgraphstruc- ture. Neurocomputing 619, 129135

  58. [66]

    Hfa- net: hierarchical feature aggregation network for micro-expression recognition

    Zhang, M., Yang, W., Wang, L., Wu, Z., Chen, D., 2025c. Hfa- net: hierarchical feature aggregation network for micro-expression recognition. Complex & Intelligent Systems 11, 169

  59. [67]

    Sfml-net: Spatial feature-guided motion learning network for micro- expression recognition

    Zhang, M., Zhu, T., Gong, H., Yang, W., Wei, F., Yao, L., 2026a. Sfml-net: Spatial feature-guided motion learning network for micro- expression recognition. Journal of King Saud University Computer and Information Sciences

  60. [68]

    Fast micro-expression recognition method based on bi-directional optical flow

    Zhang, Y., Fei, Z., Zhou, W., Fei, M., 2025d. Fast micro-expression recognition method based on bi-directional optical flow. Applied Intelligence 55, 865

  61. [69]

    Lrcn- stan: micro-expression recognition via dual-stream lrcn-enhanced spatiotemporal attention network

    Zhang, Y., Guo, X., Wu, S., Xu, W., Zheng, Y., 2026b. Lrcn- stan: micro-expression recognition via dual-stream lrcn-enhanced spatiotemporal attention network. International Journal of Machine Learning and Cybernetics 17, 54

  62. [70]

    Zhang, Y., Wang, H., Xu, Y., Mao, X., Xu, T., Zhao, S., Chen, E.,

  63. [71]

    1391–1396

    Adaptive graph attention network with temporal fusion for micro-expressions recognition, in: IEEE International Conference on Multimedia and Expo (ICME), pp. 1391–1396

  64. [72]

    Zhang, Z., Tang, W., Chen, H., 2025e. Rethinking key-frame- based micro-expression recognition: A robust and accurate frame- work against key-frame errors, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 12274–12283

  65. [73]

    Dfme: A new benchmark for dynamic facial micro- expression recognition

    Zhao, S., Tang, H., Mao, X., Liu, S., Zhang, Y., Wang, H., Xu, T., Chen, E., 2023. Dfme: A new benchmark for dynamic facial micro- expression recognition. IEEE Transactions on Affective Computing 15, 1371–1386

    2023

  66. [74]

    Sta-gcn: Spatio-temporal au graph convolution network for facial micro-expression recognition, in:ChineseConferenceonPatternRecognitionandComputerVision, Springer

    Zhao, X., Ma, H., Wang, R., 2021. Sta-gcn: Spatio-temporal au graph convolution network for facial micro-expression recognition, in:ChineseConferenceonPatternRecognitionandComputerVision, Springer. pp. 80–91

    2021

  67. [75]

    Featurere- finement: An expression-specific feature learning and fusion method for micro-expression recognition

    Zhou,L.,Mao,Q.,Huang,X.,Zhang,F.,Zhang,Z.,2022. Featurere- finement: An expression-specific feature learning and fusion method for micro-expression recognition. Pattern Recognition 122, 108275. Miss. Nandani Sharma is a doctoral candidate at IITMandiintheSCEEdepartment,working...

    2022

Pith tools

Reviewed June 29, 2026 · model on record in the stance chip above.