pith. sign in

arxiv: 2604.06190 · v1 · submitted 2026-03-02 · 💻 cs.HC

SASLO: A Scene-Aware Spatial Layout Optimization System for AR-SSVEP

Beining Cao , Xiaowei Jiang , Charlie Li-Ting Tsai , Daniel Leong , Thomas Do , Chin-Teng Lin This is my paper

Pith reviewed 2026-05-15 18:32 UTC · model grok-4.3

classification 💻 cs.HC
keywords AR-SSVEPspatial layout optimizationscene luminancebrain-computer interfacelinear contextual banditaugmented realityoutdoor BCI
0
0 comments X

The pith

A scene-aware system adapts AR-SSVEP stimulus layouts to real-world luminance and spacing for stronger brain signals outdoors.

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

The paper presents SASLO, a system that jointly optimizes where to place visual stimuli in augmented-reality brain-computer interfaces by accounting for local scene brightness and the distance between stimuli. It estimates luminance from ordinary RGB images and feeds that context into a linear bandit model whose recommendations are trained on data from simple pilot tests. In a ten-person outdoor trial the adaptive layouts produced 0.89 average accuracy and 35.74 bits per minute information transfer rate over three-second windows, beating two non-adaptive baselines.

Core claim

The SASLO system estimates scene luminance with an RGB-CIE method and uses a linear contextual bandit model, trained on rewards collected from two single-factor pilot experiments, to recommend joint luminance-and-inter-stimulus-distance layouts that improve SSVEP elicitation; an outdoor online experiment with ten subjects confirms average accuracy of 0.89 and information transfer rate of 35.74 bits/min with a 3 s window while outperforming fixed baselines.

What carries the argument

Linear contextual bandit model that takes estimated scene luminance as context and outputs recommended spatial layouts of AR stimuli.

If this is right

  • Real-time luminance estimation plus bandit selection can keep SSVEP performance stable as users move through changing outdoor light.
  • Joint optimization of brightness and spacing yields higher accuracy and information rate than optimizing either factor alone.
  • The same pipeline supports online recalibration when the user enters a new scene without requiring full retraining.
  • AR-SSVEP interfaces become practical for mobile use once layout decisions adapt to the immediate visual surround.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach could be extended to indoor environments with varying artificial lighting or to other AR visual evoked-potential paradigms.
  • Pre-computing a small library of scene prototypes might further reduce the number of pilot trials needed for new users.
  • Combining the bandit with live camera feeds on lightweight AR glasses would allow continuous layout updates during natural movement.

Load-bearing premise

The linear bandit model trained on a few pilot scenes will still pick good layouts when it encounters completely new outdoor environments.

What would settle it

An outdoor test in a high-contrast or low-light scene where the SASLO-chosen layout produces accuracy below 0.7 or ITR below 20 bits/min while a fixed grid layout performs better.

Figures

Figures reproduced from arXiv: 2604.06190 by Beining Cao, Charlie Li-Ting Tsai, Chin-Teng Lin, Daniel Leong, Thomas Do, Xiaowei Jiang.

Figure 1
Figure 1. Figure 1: Overview of the SASLO system architecture and module interactions. In the diagram, solid black arrows represent data [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Diagram of the single-input fuzzy rule-based model. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Illustration of the offline simulation paradigm and ex [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Results of the six-class offline experiment. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Overview of the outdoor online AR-SSVEP experiment. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FFT-based amplitude spectra at 7.5 Hz across different methods. Single-sided FFT spectra averaged across subjects are [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison of rotation-related power differences under [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
read the original abstract

Steady-state visual evoked potential (SSVEP) is widely used in brain-computer interfaces (BCIs) due to its reliability. With the integration of augmented reality (AR), AR-SSVEP enables more intuitive interaction by embedding visual stimuli into real-world environments. However, unlike conventional computer screen-based SSVEP (CS-SSVEP) systems with stable visual conditions, AR-SSVEP performance is influenced by real-world scene factors, such as luminance and color, which degrade stimulus perception and weaken SSVEP elicitation. Nevertheless, existing studies primarily focus on offline analyses of SSVEP-related factors in indoor settings, while online adaptive optimization for outdoor AR-SSVEP remains limited. Therefore, a scenario-aware spatial layout optimization (SASLO) system for AR-SSVEP is proposed, which jointly considers scene luminance and inter-stimulus distance (ISD) for adaptive stimulus layout optimization. Scene luminance is estimated using an RGB-CIE based method, and the extracted context is incorporated into a linear contextual bandit (LCB) model to recommend optimized spatial layouts. Two pilot single-factor experiments are conducted to characterize the effects of luminance and ISD on SSVEP performance and to construct reliable rewards for model training. An outdoor online experiment with ten subjects further validates the proposed joint optimization method, achieving an average accuracy of 0.89 and an information transfer rate of 35.74 bits/min with a 3 s input window, and consistently outperforming two baseline methods. Overall, the proposed SASLO system is shown to improve the robustness of AR-SSVEP in real-world outdoor environments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper proposes SASLO, a scene-aware spatial layout optimization system for AR-SSVEP BCIs. It estimates outdoor scene luminance via an RGB-CIE method, feeds the context into a linear contextual bandit (LCB) model whose rewards are derived from two single-factor pilot experiments (luminance and inter-stimulus distance), and recommends optimized stimulus layouts. An outdoor online validation with 10 subjects reports average accuracy 0.89 and ITR 35.74 bits/min (3 s window), outperforming two unspecified baselines.

Significance. If the generalization claim holds, the work would be a useful step toward robust outdoor AR-BCIs by providing an adaptive, data-driven layout optimizer that accounts for real-world luminance and spacing. The separation of pilot reward collection from the outdoor test keeps circularity low and supplies a concrete, falsifiable performance target (0.89 accuracy, 35.74 bits/min).

major comments (3)
  1. [Outdoor validation experiment] The outdoor online experiment section provides no statistical tests (p-values, confidence intervals, or effect sizes) or per-subject raw data to support the claim of consistent outperformance over the two baselines; with only 10 subjects this omission makes it impossible to judge whether the reported 0.89 accuracy and 35.74 bits/min ITR are reliably superior.
  2. [LCB model and reward construction] The LCB is trained exclusively on rewards from two separate single-factor pilots; no cross-scene hold-out, leave-one-scene-out, or joint-factor ablation is reported to test whether the linear model captures luminance–ISD interactions or transfers to unseen outdoor scenes whose statistics differ from the pilots.
  3. [Experimental comparison] Baseline methods are mentioned only as “two baseline methods” without implementation details, parameter settings, or justification for their selection; this prevents assessment of whether the reported gains are due to the joint optimization or to weak comparators.
minor comments (2)
  1. [Scene luminance estimation] The RGB-CIE luminance estimator is introduced without a quantitative validation against ground-truth luminance measurements in the target outdoor scenes.
  2. [Method] Notation for inter-stimulus distance (ISD) and the exact form of the context vector supplied to the LCB should be defined explicitly in a table or equation.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major point below and will revise the paper where appropriate to strengthen the presentation.

read point-by-point responses

  1. Referee: [Outdoor validation experiment] The outdoor online experiment section provides no statistical tests (p-values, confidence intervals, or effect sizes) or per-subject raw data to support the claim of consistent outperformance over the two baselines; with only 10 subjects this omission makes it impossible to judge whether the reported 0.89 accuracy and 35.74 bits/min ITR are reliably superior.

    Authors: We agree that statistical tests and per-subject data are necessary to support performance claims. In the revised manuscript we will add paired statistical comparisons (t-tests or Wilcoxon signed-rank tests as appropriate) between SASLO and each baseline, reporting p-values, 95% confidence intervals, and effect sizes. We will also include a supplementary table listing per-subject accuracy and ITR values from the outdoor experiment so readers can evaluate consistency across the n=10 participants. revision: yes

  2. Referee: [LCB model and reward construction] The LCB is trained exclusively on rewards from two separate single-factor pilots; no cross-scene hold-out, leave-one-scene-out, or joint-factor ablation is reported to test whether the linear model captures luminance–ISD interactions or transfers to unseen outdoor scenes whose statistics differ from the pilots.

    Authors: The two pilot studies were deliberately single-factor to obtain independent empirical reward functions for luminance and ISD before combining them linearly inside the contextual bandit. The outdoor online experiment with ten subjects in previously unseen real-world scenes already serves as an external validation of transfer. We acknowledge that explicit tests for luminance–ISD interactions or leave-one-scene-out cross-validation on the pilot data are absent. In revision we will add a limitations paragraph discussing this modeling choice and outlining future joint-factor ablation experiments. revision: partial

  3. Referee: [Experimental comparison] Baseline methods are mentioned only as “two baseline methods” without implementation details, parameter settings, or justification for their selection; this prevents assessment of whether the reported gains are due to the joint optimization or to weak comparators.

    Authors: We will expand the experimental section to fully describe the two baselines: (1) a fixed-layout condition using standard SSVEP stimulus spacing and size, and (2) a random-layout condition that samples valid positions within the AR scene while respecting minimum distance constraints. We will report all parameter values (stimulus size, frequency, color, etc.) and justify the baselines as representative non-adaptive controls. This will make clear that performance differences arise from the scene-aware joint optimization rather than from weak comparators. revision: yes

Circularity Check

0 steps flagged

No significant circularity; validation uses independent outdoor data

full rationale

The paper constructs rewards for the linear contextual bandit from two separate single-factor pilot experiments (luminance and ISD) and then evaluates the resulting joint optimization policy in a distinct outdoor online experiment with ten subjects. This produces reported metrics (accuracy 0.89, ITR 35.74 bits/min) on held-out scenes and subjects. No equation reduces to its own fitted parameters by construction, no self-citation supplies a load-bearing uniqueness theorem, and the central performance claim rests on fresh empirical measurements rather than re-deriving the pilot inputs. The derivation chain therefore remains self-contained.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The approach depends on empirical data from controlled pilot studies to parameterize the adaptive model and assumes these relationships hold in dynamic outdoor environments.

free parameters (1)
  • Linear contextual bandit parameters
    The LCB model parameters are fitted using rewards derived from the pilot single-factor experiments.
axioms (2)
  • domain assumption The RGB-CIE based method provides an accurate estimate of scene luminance relevant to SSVEP perception
    This estimation is used to incorporate context into the optimization model.
  • domain assumption The effects of luminance and ISD on SSVEP performance observed in pilot experiments can be used to define reliable rewards for the LCB model
    These rewards train the model for adaptive layout recommendations.

pith-pipeline@v0.9.0 · 5609 in / 1491 out tokens · 69183 ms · 2026-05-15T18:32:44.798871+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

45 extracted references · 45 canonical work pages · 1 internal anchor

  1. [1]

    Attention- based multiscale tcnn for ssvep classification and its application to bionic intelligent soft gripper control,

    R. Wei, C. Hua, J. Chen, D. Mu, and J. Zhao, “Attention- based multiscale tcnn for ssvep classification and its application to bionic intelligent soft gripper control,”IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2025

    work page 2025

  2. [2]

    Ssvep-based brain–computer interface controlled functional electrical stimulation system for upper extremity rehabilitation,

    X. Zhao, Y . Chu, J. Han, and Z. Zhang, “Ssvep-based brain–computer interface controlled functional electrical stimulation system for upper extremity rehabilitation,” IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 46, no. 7, pp. 947–956, 2016

    work page 2016

  3. [3]

    Stimulus-stimulus transfer based on time- frequency-joint representation in ssvep-based bcis,

    Z. Wang, C. M. Wong, A. Rosa, T. Qian, T.-P. Jung, and F. Wan, “Stimulus-stimulus transfer based on time- frequency-joint representation in ssvep-based bcis,”IEEE Transactions on Biomedical Engineering, vol. 70, no. 2, pp. 603–615, 2022

    work page 2022

  4. [4]

    Emd-fuzzy: An empirical mode decomposi- tion based fuzzy model for cross-stimulus transfer learning of ssvep,

    B. Cao, X. Jiang, D. Leong, C. L.-T. Tsai, Y .-C. Chang, T. Doet al., “Emd-fuzzy: An empirical mode decomposi- tion based fuzzy model for cross-stimulus transfer learning of ssvep,”arXiv preprint arXiv:2501.17475, 2025

    work page arXiv 2025

  5. [5]

    Direct-sense brain–computer interfaces and wearable computers,

    C.-T. Lin and T.-T. N. Do, “Direct-sense brain–computer interfaces and wearable computers,”IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 51, no. 1, pp. 298–312, 2020

    work page 2020

  6. [6]

    Optimization of stimulus properties for ssvep- based bmi system with a heads-up display to control in-vehicle features,

    H. Hamidi Shishavan, R. Roy, K. Golzari, A. Singla, D. Zalozhin, D. Lohan, M. Farooq, E. M. Dede, and I. Kim, “Optimization of stimulus properties for ssvep- based bmi system with a heads-up display to control in-vehicle features,”Plos one, vol. 19, no. 9, p. e0308506, 2024

    work page 2024

  7. [7]

    An ssvep-based bci with leds visual stimuli using dynamic window cca algorithm,

    C. Liu, M. Duan, Z. Duan, A. Liu, Z. Lu, and H. Wang, “An ssvep-based bci with leds visual stimuli using dynamic window cca algorithm,”Biomedical Signal Processing and Control, vol. 76, p. 103727, 2022

    work page 2022

  8. [8]

    Visual stimulus color effect on ssvep- bci in augmented reality,

    Y . Du and X. Zhao, “Visual stimulus color effect on ssvep- bci in augmented reality,”Biomedical Signal Processing and Control, vol. 78, p. 103906, 2022

    work page 2022

  9. [9]

    Dataset of binocularly coded steady-state visual evoked potentials recorded with an augmented reality headset,

    Y . Ke, Y . Han, P. Liu, and D. Ming, “Dataset of binocularly coded steady-state visual evoked potentials recorded with an augmented reality headset,”Scientific Data, vol. 12, no. 1, p. 1338, 2025

    work page 2025

  10. [10]

    Brain- controlled, ar-based home automation system using ssvep- based brain-computer interface and eog-based eye tracker: A feasibility study for the elderly end user,

    S. Park, J. Ha, J. Park, K. Lee, and C.-H. Im, “Brain- controlled, ar-based home automation system using ssvep- based brain-computer interface and eog-based eye tracker: A feasibility study for the elderly end user,”IEEE Transac- tions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 544–553, 2022

    work page 2022

  11. [11]

    Towards bci-based interfaces for augmented reality: feasibility, design and evaluation,

    H. Si-Mohammed, J. Petit, C. Jeunet, F. Argelaguet, F. Spindler, A. Evain, N. Roussel, G. Casiez, and A. Lécuyer, “Towards bci-based interfaces for augmented reality: feasibility, design and evaluation,”IEEE trans- actions on visualization and computer graphics, vol. 26, no. 3, pp. 1608–1621, 2018

    work page 2018

  12. [12]

    Enhanced system robustness of asynchronous bci in augmented reality using steady-state motion visual evoked potential,

    A. Ravi, J. Lu, S. Pearce, and N. Jiang, “Enhanced system robustness of asynchronous bci in augmented reality using steady-state motion visual evoked potential,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 30, pp. 85–95, 2022

    work page 2022

  13. [13]

    Ambient brightness issues in ar- bci applications: Phenomenon and solutions,

    R. Zhang and D. Yao, “Ambient brightness issues in ar- bci applications: Phenomenon and solutions,”Authorea Preprints, 2022

    work page 2022

  14. [14]

    Estimating the cognitive load in physical spatial naviga- tion,

    T.-T. N. Do, A. K. Singh, C. A. T. Cortes, and C.-T. Lin, “Estimating the cognitive load in physical spatial naviga- tion,” in2020 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE, 2020, pp. 568–575

    work page 2020

  15. [15]

    Improving ar-ssvep recognition accuracy under high ambient brightness through iterative learning,

    R. Zhang, L. Cao, Z. Xu, Y . Zhang, L. Zhang, Y . Hu, M. Chen, and D. Yao, “Improving ar-ssvep recognition accuracy under high ambient brightness through iterative learning,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 1796–1806, 2023

    work page 2023

  16. [16]

    Performance en- hancement of an ssvep-based brain–computer interface in augmented reality through adaptive color adjustment of visual stimuli for optimal background contrast,

    C.-U. Kim, S. Park, and C.-H. Im, “Performance en- hancement of an ssvep-based brain–computer interface in augmented reality through adaptive color adjustment of visual stimuli for optimal background contrast,”IEEE Transactions on Neural Systems and Rehabilitation Engi- neering, 2025

    work page 2025

  17. [17]

    Retrosplenial segregation reflects the navigation load during ambulatory 12 movement,

    T.-T. N. Do, T.-P. Jung, and C.-T. Lin, “Retrosplenial segregation reflects the navigation load during ambulatory 12 movement,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 29, pp. 488–496, 2021

    work page 2021

  18. [18]

    Wearable brain–computer interfaces based on steady-state visually evoked potentials and augmented reality: A review,

    L. Angrisani, P. Arpaia, E. De Benedetto, L. Duraccio, F. L. Regio, and A. Tedesco, “Wearable brain–computer interfaces based on steady-state visually evoked potentials and augmented reality: A review,”IEEE Sensors Journal, vol. 23, no. 15, pp. 16 501–16 514, 2023

    work page 2023

  19. [19]

    Visual stimulus background effects on ssvep-based bci towards a practical robot car control,

    X. Shu, L. Yao, J. Meng, X. Sheng, and X. Zhu, “Visual stimulus background effects on ssvep-based bci towards a practical robot car control,”International Journal of Humanoid Robotics, vol. 12, no. 02, p. 1550014, 2015

    work page 2015

  20. [20]

    Optimal stimulus properties for steady-state visually evoked potential brain–computer interfaces: a scoping review,

    C. Reitelbach and K. Oyibo, “Optimal stimulus properties for steady-state visually evoked potential brain–computer interfaces: a scoping review,”Multimodal Technologies and Interaction, vol. 8, no. 2, p. 6, 2024

    work page 2024

  21. [21]

    User-specific channel selection method to improve ssvep bci decoding robustness against variable inter-stimulus distance,

    A. Ravi, S. Pearce, X. Zhang, and N. Jiang, “User-specific channel selection method to improve ssvep bci decoding robustness against variable inter-stimulus distance,” in 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER). IEEE, 2019, pp. 283–286

    work page 2019

  22. [22]

    Ssvep stimulus layout effect on accuracy of brain-computer interfaces in augmented reality glasses,

    X. Zhao, C. Liu, Z. Xu, L. Zhang, and R. Zhang, “Ssvep stimulus layout effect on accuracy of brain-computer interfaces in augmented reality glasses,”IEEE Access, vol. 8, pp. 5990–5998, 2020

    work page 2020

  23. [23]

    Effect of competing stimuli on ssvep-based bci,

    K. B. Ng, A. P. Bradley, and R. Cunnington, “Effect of competing stimuli on ssvep-based bci,” in2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE, 2011, pp. 6307– 6310

    work page 2011

  24. [24]

    M. D. Fairchild,Color appearance models. John Wiley & Sons, 2013

    work page 2013

  25. [25]

    Colour spaces,

    D. Briggs, “Colour spaces,” inThe Routledge Handbook of Philosophy of Colour. Routledge, 2020, pp. 140–156

    work page 2020

  26. [26]

    Retinexformer: One-stage retinex-based trans- former for low-light image enhancement,

    Y . Cai, H. Bian, J. Lin, H. Wang, R. Timofte, and Y . Zhang, “Retinexformer: One-stage retinex-based trans- former for low-light image enhancement,” inProceedings of the IEEE/CVF international conference on computer vision, 2023, pp. 12 504–12 513

    work page 2023

  27. [27]

    Beyond the pixel: a photometrically calibrated hdr dataset for luminance and color prediction,

    C. Bolduc, J. Giroux, M. Hébert, C. Demers, and J.-F. Lalonde, “Beyond the pixel: a photometrically calibrated hdr dataset for luminance and color prediction,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 8071–8081

    work page 2023

  28. [28]

    A contextual-bandit approach to personalized news article recommendation,

    L. Li, W. Chu, J. Langford, and R. E. Schapire, “A contextual-bandit approach to personalized news article recommendation,” inProceedings of the 19th international conference on World wide web, 2010, pp. 661–670

    work page 2010

  29. [29]

    Linear contextual bandits with knapsacks,

    S. Agrawal and N. Devanur, “Linear contextual bandits with knapsacks,”Advances in neural information process- ing systems, vol. 29, 2016

    work page 2016

  30. [30]

    Neural contextual bandits with ucb-based exploration,

    D. Zhou, L. Li, and Q. Gu, “Neural contextual bandits with ucb-based exploration,” inInternational conference on machine learning. PMLR, 2020, pp. 11 492–11 502

    work page 2020

  31. [31]

    An ssvep stimuli design using real-time camera view with object recognition,

    S.-K. Chen, C.-S. Chen, Y .-K. Wang, and C.-T. Lin, “An ssvep stimuli design using real-time camera view with object recognition,” in2020 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE, 2020, pp. 562–567

    work page 2020

  32. [32]

    A novel 3d paradigm for target expansion of augmented reality ssvep,

    B. Cao, C. L.-T. Tsai, N. Zhou, T. Do, and C.-T. Lin, “A novel 3d paradigm for target expansion of augmented reality ssvep,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2025

    work page 2025

  33. [33]

    Independent component analysis for eeg source localization,

    L. Zhukov, D. Weinstein, and C. Johnson, “Independent component analysis for eeg source localization,”IEEE Engineering in Medicine and Biology Magazine, vol. 19, no. 3, pp. 87–96, 2002

    work page 2002

  34. [34]

    Lin and C

    C.-T. Lin and C. G. Lee,Neural fuzzy systems: a neuro- fuzzy synergism to intelligent systems. Prentice-Hall, Inc., 1996

    work page 1996

  35. [35]

    Interpretable dual-filter fuzzy neural networks for affective brain-computer interfaces,

    X. Jiang, Y . Chen, N. R. Pal, Y .-C. Chang, Y . Yang, T. Do, and C.-T. Lin, “Interpretable dual-filter fuzzy neural networks for affective brain-computer interfaces,”arXiv preprint arXiv:2502.17445, 2025

    work page arXiv 2025

  36. [36]

    ifuzzytl: Interpretable fuzzy transfer learning for steady-state visual evoked potentials brain–computer interfaces system,

    X. Jiang, B. Cao, L. Ou, Y .-C. Chang, T. Do, and C.- T. Lin, “ifuzzytl: Interpretable fuzzy transfer learning for steady-state visual evoked potentials brain–computer interfaces system,”IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2025

    work page 2025

  37. [37]

    A fuzzy logic-based approach to predict hu- man interaction by functional near-infrared spectroscopy,

    X. Jiang, L. Ou, Y . Chen, N. Ao, Y .-C. Chang, T. Do, and C.-T. Lin, “A fuzzy logic-based approach to predict hu- man interaction by functional near-infrared spectroscopy,” IEEE Transactions on Fuzzy Systems, 2025

    work page 2025

  38. [38]

    Tailoring resonant modes in dual cavities for transmissive structural colors with high brightness and high purity,

    I. Jung, H. Kim, S. Ju, D. Kim, H. Kwak, H. Wang, L. Jay Guo, H. J. Park, J. G. Ok, and K.-T. Lee, “Tailoring resonant modes in dual cavities for transmissive structural colors with high brightness and high purity,”Optics Express, vol. 32, no. 15, pp. 26 136–26 146, 2024

    work page 2024

  39. [39]

    Can we estimate which colors our participants see? com- paring results from different gamma correction methods,

    D. Epicoco, D. Jonauskaite, C. Mohr, and C. A. Parraga, “Can we estimate which colors our participants see? com- paring results from different gamma correction methods,” i-Perception, vol. 15, no. 5, p. 20416695241278562, 2024

    work page 2024

  40. [40]

    The cityscapes dataset for semantic urban scene under- standing,

    M. Cordts, M. Omran, S. Ramos, T. Rehfeld, M. En- zweiler, R. Benenson, U. Franke, S. Roth, and B. Schiele, “The cityscapes dataset for semantic urban scene under- standing,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 3213– 3223

    work page 2016

  41. [41]

    Active semi- supervision for pairwise constrained clustering,

    S. Basu, A. Banerjee, and R. J. Mooney, “Active semi- supervision for pairwise constrained clustering,” inPro- ceedings of the 2004 SIAM international conference on data mining. SIAM, 2004, pp. 333–344

    work page 2004

  42. [42]

    Visual perception of extent and the geometry of visual space,

    J. M. Foley, N. P. Ribeiro-Filho, and J. A. Da Silva, “Visual perception of extent and the geometry of visual space,”Vision Research, vol. 44, no. 2, pp. 147–156, 2004

    work page 2004

  43. [43]

    Restricting the vertical and horizontal extent of the field-of-view: effects on manoeuvring performance,

    S. E. M. Jansen, A. Toet, and N. J. Delleman, “Restricting the vertical and horizontal extent of the field-of-view: effects on manoeuvring performance,”The Ergonomics Open Journal, vol. 3, no. 1, pp. 19–24, 2010

    work page 2010

  44. [44]

    Adam: A Method for Stochastic Optimization

    D. P. Kingma, “Adam: A method for stochastic optimiza- tion,”arXiv preprint arXiv:1412.6980, 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  45. [45]

    Transferring subject-specific knowledge across stimulus frequencies in ssvep-based bcis,

    C. M. Wong, Z. Wang, A. C. Rosa, C. P. Chen, T.-P. Jung, Y . Hu, and F. Wan, “Transferring subject-specific knowledge across stimulus frequencies in ssvep-based bcis,”IEEE Transactions on Automation Science and Engineering, vol. 18, no. 2, pp. 552–563, 2021

    work page 2021