EyeMakeYou: Identity-, Task-, and Subjective-State-Conditioned Diffusion for High-Frequency Gaze Synthesis

• URL: http://arxiv.org/abs/2609.04501v1
• Date: Thu, 03 Sep 2026 21:41:30 GMT
• Title: EyeMakeYou: Identity-, Task-, and Subjective-State-Conditioned Diffusion for High-Frequency Gaze Synthesis
• Abstract summary: Eye movement biometrics (EMB) is an emerging behavioral modality for user authentication.<n>EMB systems require diverse, high-quality gaze recordings that are expensive to collect and often unavailable at the scale needed for model development.<n>We propose EyeMakeYou, a multi-conditional denoising diffusion framework for subject-specific, high-frequency gaze synthesis.
• Score: 3.0755265578451874
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Eye movement biometrics (EMB) is an emerging behavioral modality for user authentication, particularly in virtual- and augmented-reality systems, where gaze dynamics contain distinctive subject-specific features. However, robust EMB systems require diverse, high-quality gaze recordings that are expensive to collect and often unavailable at the scale needed for model development. Generative models can mitigate data scarcity, but existing methods either synthesize generic gaze behavior or personalize signals primarily by identity, without jointly representing the user's task and subjective state. Consequently, generated signals may appear visually realistic while failing to retain the behavioral properties required for biometric applications. To address this limitation, we propose EyeMakeYou, a multi-conditional denoising diffusion framework for subject-specific, high-frequency gaze synthesis. EyeMakeYou generates 5-s, 1000-Hz bivariate gaze-velocity sequences from an identity-removed reference trajectory and conditions the denoising process on an identity embedding, a task embedding, and self-reported ratings of overall difficulty, mental tiredness, and eye tiredness. Its objective combines diffusion noise prediction and identity preservation with multi-resolution spectral, drift-consistency, and event-weighted local-smoothness losses. Experiments on GazeBase show that EyeMakeYou achieves higher median spatial accuracy and greater real--synthetic similarity in the embedding feature space than the existing generative approaches, while retaining selected task-dependent associations between subjective reports and oculomotor features. These findings support conditional diffusion as a practical approach for augmenting gaze datasets for biometric and interactive applications.

Related papers

• FVeinSyn: Synthetic Finger Vein Image Generator [75.79003800700403]
  A major challenge in finger vein recognition is the lack of large-scale public datasets.<n>We propose FVeinSyn, a large-scale controllable synthetic data generation framework for finger vein.
  arXiv  Detail & Related papers  (2026-08-27T14:46:43Z)
• A Dual-Stream Challenge-Response Protocol for Ocular Liveness Verification [0.0]
  Ocular biometric systems face sophisticated presentation attacks, including high-resolution video replays and real-time generative deepfakes.<n>This paper proposes a dual-stream challenge-response framework for ocular liveness verification.<n>We show that a multi-round challenge design improves the detection of generative deepfakes when a non-zero-latency rendering gap exists.
  arXiv  Detail & Related papers  (2026-07-10T18:15:23Z)
• Enhancing Eye Movement Biometrics for User Authentication via Continuous Gaze Offset Score Fusion [2.2557806157585834]
  Eye movement biometrics (EMB) use subject-specific gaze dynamics for user authentication and identification.<n>Recent deep learning-based EMB systems achieve strong performance by modeling temporal eye movement behavior.<n>This work examines whether continuous gaze offset can improve biometric performance when combined with existing biometric features.
  arXiv  Detail & Related papers  (2026-05-07T18:12:20Z)
• Real-time Appearance-based Gaze Estimation for Open Domains [15.195617974731766]
  We propose a robust AGE framework that enhances generalization without requiring additional human-annotated data.<n>To mitigate the impact of anisotropic inter-dataset label deviation, we reformulate gaze regression as a multi-task learning problem.<n>Our MobileNet-based lightweight model generalization achieves performance competitive with the state-of-the-art (SOTA) UniGaze-H.
  arXiv  Detail & Related papers  (2026-03-27T19:33:18Z)
• FAIM: Frequency-Aware Interactive Mamba for Time Series Classification [87.84511960413715]
  Time series classification (TSC) is crucial in numerous real-world applications, such as environmental monitoring, medical diagnosis, and posture recognition.<n>We propose FAIM, a lightweight Frequency-Aware Interactive Mamba model.<n>We show that FAIM consistently outperforms existing state-of-the-art (SOTA) methods, achieving a superior trade-off between accuracy and efficiency.
  arXiv  Detail & Related papers  (2025-11-26T08:36:33Z)
• GaitCrafter: Diffusion Model for Biometric Preserving Gait Synthesis [14.174192604480599]
  GaitCrafter is a diffusion-based framework for synthesizing realistic gait sequences in the silhouette domain.<n>Our approach enables the generation of temporally consistent and identity-preserving gait sequences.<n>We introduce a mechanism to generate novel identities-synthetic individuals not present in the original dataset.
  arXiv  Detail & Related papers  (2025-08-18T18:32:42Z)
• Generate Aligned Anomaly: Region-Guided Few-Shot Anomaly Image-Mask Pair Synthesis for Industrial Inspection [53.137651284042434]
  Anomaly inspection plays a vital role in industrial manufacturing, but the scarcity of anomaly samples limits the effectiveness of existing methods.<n>We propose Generate grained Anomaly (GAA), a region-guided, few-shot anomaly image-mask pair generation framework.<n>GAA generates realistic, diverse, and semantically aligned anomalies using only a small number of samples.
  arXiv  Detail & Related papers  (2025-07-13T12:56:59Z)
• Multi-focal Conditioned Latent Diffusion for Person Image Synthesis [59.113899155476005]
  The Latent Diffusion Model (LDM) has demonstrated strong capabilities in high-resolution image generation.<n>We propose a Multi-focal Conditioned Latent Diffusion (MCLD) method to address these limitations.<n>Our approach utilizes a multi-focal condition aggregation module, which effectively integrates facial identity and texture-specific information.
  arXiv  Detail & Related papers  (2025-03-19T20:50:10Z)
• Synthetic Forehead-creases Biometric Generation for Reliable User Verification [6.639785884921617]
  We present a new framework to synthesize forehead-crease image data while maintaining important features, such as uniqueness and realism. We evaluate the diversity and realism of the generated forehead-crease images using the Fr'echet Inception Distance (FID) and the Structural Similarity Index Measure (SSIM)
  arXiv  Detail & Related papers  (2024-08-28T10:33:00Z)
• Multimodal Adaptive Fusion of Face and Gait Features using Keyless attention based Deep Neural Networks for Human Identification [67.64124512185087]
  Soft biometrics such as gait are widely used with face in surveillance tasks like person recognition and re-identification. We propose a novel adaptive multi-biometric fusion strategy for the dynamic incorporation of gait and face biometric cues by leveraging keyless attention deep neural networks.
  arXiv  Detail & Related papers  (2023-03-24T05:28:35Z)
• Evidential Sparsification of Multimodal Latent Spaces in Conditional Variational Autoencoders [63.46738617561255]
  We consider the problem of sparsifying the discrete latent space of a trained conditional variational autoencoder. We use evidential theory to identify the latent classes that receive direct evidence from a particular input condition and filter out those that do not. Experiments on diverse tasks, such as image generation and human behavior prediction, demonstrate the effectiveness of our proposed technique.
  arXiv  Detail & Related papers  (2020-10-19T01:27:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.