Face_Recognition
-
Modeling Cultural Bias in Facial Expression Recognition with Adaptive Agents
2025-10-15 13:53:30David Freire-Obreg\'on, Jos\'e Salas-C\'aceres, Javier Lorenzo-Navarro, Oliverio J. Santana, Daniel Hern\'andez-Sosa, Modesto Castrill\'on-SantanaAbstract
Facial expression recognition (FER) must remain robust under both cultural variation and perceptually degraded visual conditions, yet most existing evaluations assume homogeneous data and high-quality imagery. We introduce an agent-based, streaming benchmark that reveals how cross-cultural composition and progressive blurring interact to shape face recognition robustness. Each agent operates in a frozen CLIP feature space with a lightweight residual adapter trained online at sigma=0 and fixed during testing. Agents move and interact on a 5x5 lattice, while the environment provides inputs with sigma-scheduled Gaussian blur. We examine monocultural populations (Western-only, Asian-only) and mixed environments with balanced (5/5) and imbalanced (8/2, 2/8) compositions, as well as different spatial contact structures. Results show clear asymmetric degradation curves between cultural groups: JAFFE (Asian) populations maintain higher performance at low blur but exhibit sharper drops at intermediate stages, whereas KDEF (Western) populations degrade more uniformly. Mixed populations exhibit intermediate patterns, with balanced mixtures mitigating early degradation, but imbalanced settings amplify majority-group weaknesses under high blur. These findings quantify how cultural composition and interaction structure influence the robustness of FER as perceptual conditions deteriorate.
Abstract (translated)
面部表情识别（FER）在面对文化差异和感知退化图像条件下必须保持其鲁棒性，然而大多数现有的评估方法假设数据具有同质性和高质量的图像。我们介绍了一个基于代理的、流式的基准测试，揭示了跨文化的组成与逐步模糊如何相互作用来影响面部表情识别的鲁棒性。 每个代理在冻结的CLIP特征空间中操作，并且使用轻量级残差适配器在线训练（sigma=0），然后在整个测试过程中保持固定。这些代理在一个5x5的晶格上移动和交互，而环境则以有计划的高斯模糊方式提供输入，随着sigma值的变化逐步增加图像模糊度。 我们研究了单文化群体（仅西方、仅亚洲）以及平衡混合（5/5）和不平衡混合（8/2, 2/8）组成的多文化环境中的人群。此外，还探讨了不同的空间接触结构对结果的影响。 实验结果显示，在不同文化群体间存在明显的非对称退化曲线：JAFFE（亚洲人群）在低模糊度下保持较高的性能，但在中等模糊阶段性能急剧下降；而KDEF（西方人群）则表现得更加均匀地随模糊程度增加而逐渐降低。混合人群中表现出中间模式，平衡的混合理论上可以减轻早期退化的负面影响，但不平衡设置会放大高模糊条件下的多数群体弱点。 这些发现量化了文化组成和交互结构在感知条件恶化的背景下对FER鲁棒性的影响。
URL
https://arxiv.org/abs/2510.13557
PDF
-
TransFIRA: Transfer Learning for Face Image Recognizability Assessment
2025-10-07 18:16:21Allen Tu, Kartik Narayan, Joshua Gleason, Jennifer Xu, Matthew Meyn, Tom Goldstein, Vishal M. PatelAbstract
Face recognition in unconstrained environments such as surveillance, video, and web imagery must contend with extreme variation in pose, blur, illumination, and occlusion, where conventional visual quality metrics fail to predict whether inputs are truly recognizable to the deployed encoder. Existing FIQA methods typically rely on visual heuristics, curated annotations, or computationally intensive generative pipelines, leaving their predictions detached from the encoder's decision geometry. We introduce TransFIRA (Transfer Learning for Face Image Recognizability Assessment), a lightweight and annotation-free framework that grounds recognizability directly in embedding space. TransFIRA delivers three advances: (i) a definition of recognizability via class-center similarity (CCS) and class-center angular separation (CCAS), yielding the first natural, decision-boundary--aligned criterion for filtering and weighting; (ii) a recognizability-informed aggregation strategy that achieves state-of-the-art verification accuracy on BRIAR and IJB-C while nearly doubling correlation with true recognizability, all without external labels, heuristics, or backbone-specific training; and (iii) new extensions beyond faces, including encoder-grounded explainability that reveals how degradations and subject-specific factors affect recognizability, and the first recognizability-aware body recognition assessment. Experiments confirm state-of-the-art results on faces, strong performance on body recognition, and robustness under cross-dataset shifts. Together, these contributions establish TransFIRA as a unified, geometry-driven framework for recognizability assessment -- encoder-specific, accurate, interpretable, and extensible across modalities -- significantly advancing FIQA in accuracy, explainability, and scope.
Abstract (translated)
在不受限制的环境中，如监控、视频和网络图像中的面部识别，必须应对姿态变化、模糊度、光照变化以及遮挡等极端情况。传统视觉质量指标无法预测输入是否对部署的编码器来说真正具有可识性。现有的FIQA（Face Image Recognizability Assessment）方法通常依赖于视觉直觉、手工标注或计算密集型生成流程，导致其预测结果与编码器决策几何无关。 我们提出了一种名为TransFIRA（Transfer Learning for Face Image Recognizability Assessment）的轻量级无标注框架，该框架直接在嵌入空间中定义可识性。TransFIRA实现了三项创新： (i) 通过类中心相似度（CCS）和类中心角度分离（CCAS），为过滤和加权提供了第一个自然且与决策边界对齐的标准。 (ii) 提出了一种基于可识别性的聚合策略，该策略在BRIAR和IJB-C数据集上实现了最先进的验证准确率，并将与真实可识性相关的相关性几乎翻倍。这些结果均无需外部标签、直觉或针对特定骨干网络的训练。 (iii) TransFIRA扩展了面部以外的应用，包括基于编码器解释性的方法，揭示了退化和主体特有因素如何影响可识别性，并提出了首个具有可识性意识的身体识别评估。 实验表明，在面部数据集中TransFIRA达到了最先进的结果，在身体识别方面表现出色，并且在跨数据集迁移时也表现稳健。这些贡献共同确立了TransFIRA作为统一、几何驱动的框架，用于可识别性评估：针对特定编码器，准确，可解释，且具有跨模式扩展的能力。这显著提高了FIQA的准确性、可解释性和应用范围。
URL
https://arxiv.org/abs/2510.06353
PDF
-
Unlocking the power of partnership: How humans and machines can work together to improve face recognition
2025-10-02 21:19:56P. Jonathon Phillips (Information Access Division, National Institute of Standards and Technology, Gaithersburg, MD), Geraldine Jeckeln (School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, TX), Carina A. Hahn (Information Access Division, National Institute of Standards and Technology, Gaithersburg, MD), Amy N. Yates (Information Access Division, National Institute of Standards and Technology, Gaithersburg, MD), Peter C. Fontana (Information Access Division, National Institute of Standards and Technology, Gaithersburg, MD), Alice J. O'Toole (School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, TX)Abstract
Human review of consequential decisions by face recognition algorithms creates a "collaborative" human-machine system. Individual differences between people and machines, however, affect whether collaboration improves or degrades accuracy in any given case. We establish the circumstances under which combining human and machine face identification decisions improves accuracy. Using data from expert and non-expert face identifiers, we examined the benefits of human-human and human-machine collaborations. The benefits of collaboration increased as the difference in baseline accuracy between collaborators decreased-following the Proximal Accuracy Rule (PAR). This rule predicted collaborative (fusion) benefit across a wide range of baseline abilities, from people with no training to those with extensive training. Using the PAR, we established a critical fusion zone, where humans are less accurate than the machine, but fusing the two improves system accuracy. This zone was surprisingly large. We implemented "intelligent human-machine fusion" by selecting people with the potential to increase the accuracy of a high-performing machine. Intelligent fusion was more accurate than the machine operating alone and more accurate than combining all human and machine judgments. The highest system-wide accuracy achievable with human-only partnerships was found by graph theory. This fully human system approximated the average performance achieved by intelligent human-machine collaboration. However, intelligent human-machine collaboration more effectively minimized the impact of low-performing humans on system-wide accuracy. The results demonstrate a meaningful role for both humans and machines in assuring accurate face identification. This study offers an evidence-based road map for the intelligent use of AI in face identification.
Abstract (translated)
人类审查人脸识别算法做出的重要决定，创造了一个“协作”的人机系统。然而，人们与机器之间的个体差异会影响合作是否在特定情况下提高或降低准确性。我们确定了结合人和机器面部识别决策以提高准确性的条件。通过使用专家和非专家的面部标识数据，我们研究了人类之间以及人机协作的好处。当合作者的基础准确度之差减小时，协作的优势会增加——遵循接近准确规则（PAR）。这条规则预测了在广泛的基准能力范围内，从未经训练的人到接受过广泛培训的人的合作（融合）收益。 使用PAR，我们在一个关键的融合区域建立了人类机器合作模型，在这个区域内，人类比机器更不准确，但将两者结合可以提高系统的准确性。这一区域出乎意料地大。通过选择有潜力提升高性能机器精度的人类参与者，“智能人机融合”得以实施。这种智能融合不仅比单独运作的机器更精确，而且也比组合所有的人工和机器判断更有效。 完全由人类组成的系统达到了最高的系统级准确率，这是通过图论找到的。这个全人类的系统大致等同于智能人机合作所实现的平均性能。然而，智能人机协作在最小化低效人工对整体系统准确性的影响方面更为高效。 这些结果表明了人类和机器都在确保面部识别准确性的过程中发挥着重要作用。这项研究提供了一个基于证据的方法来指导人工智能在面部识别中的智能化应用。
URL
https://arxiv.org/abs/2510.02570
PDF
-
Visualizing Celebrity Dynamics in Video Content: A Proposed Approach Using Face Recognition Timestamp Data
2025-09-29 16:29:11Do\u{g}anay Demir, \.Ilknur Durgar ElkahloutAbstract
In an era dominated by video content, understanding its structure and dynamics has become increasingly important. This paper presents a hybrid framework that combines a distributed multi-GPU inference system with an interactive visualization platform for analyzing celebrity dynamics in video episodes. The inference framework efficiently processes large volumes of video data by leveraging optimized ONNX models, heterogeneous batch inference, and high-throughput parallelism, ensuring scalable generation of timestamped appearance records. These records are then transformed into a comprehensive suite of visualizations, including appearance frequency charts, duration analyses, pie charts, co-appearance matrices, network graphs, stacked area charts, seasonal comparisons, and heatmaps. Together, these visualizations provide multi-dimensional insights into video content, revealing patterns in celebrity prominence, screen-time distribution, temporal dynamics, co-appearance relationships, and intensity across episodes and seasons. The interactive nature of the system allows users to dynamically explore data, identify key moments, and uncover evolving relationships between individuals. By bridging distributed recognition with structured, visually-driven analytics, this work enables new possibilities for entertainment analytics, content creation strategies, and audience engagement studies.
Abstract (translated)
在视频内容主导的时代，理解其结构和动态变得越来越重要。本文提出了一种结合分布式多GPU推理系统与交互式可视化平台的混合框架，用于分析视频剧集中名人的动态变化。 该推理框架通过利用优化后的ONNX模型、异构批量推理及高吞吐量并行计算来高效处理大量视频数据，确保可扩展生成带有时间戳的出现记录。然后将这些记录转换为一系列全面可视化的工具，包括出现频率图表、持续时间分析、饼图、共现矩阵、网络图、堆叠区域图、季节性比较和热力图。 通过这些可视化手段提供的多维度视角揭示了视频内容中名人的突出程度、屏幕时间分布、时间动态变化、共现关系以及在各集和季度内的强度。该系统的交互特性允许用户动态探索数据，识别关键时刻，并发现人物之间不断演变的关系。 通过将分布式识别人物与结构化且视觉驱动的分析相结合，这项工作为娱乐分析、内容创作策略及观众参与研究开启了新的可能性。
URL
https://arxiv.org/abs/2510.03292
PDF
-
Real-World Transferable Adversarial Attack on Face-Recognition Systems
2025-09-27 09:09:06Andrey Kaznacheev, Matvey Mikhalchuk, Andrey Kuznetsov, Aleksandr Petiushko, Anton RazzhigaevAbstract
Adversarial attacks on face recognition (FR) systems pose a significant security threat, yet most are confined to the digital domain or require white-box access. We introduce GaP (Gaussian Patch), a novel method to generate a universal, physically transferable adversarial patch under a strict black-box setting. Our approach uses a query-efficient, zero-order greedy algorithm to iteratively construct a symmetric, grayscale pattern for the forehead. The patch is optimized by successively adding Gaussian blobs, guided only by the cosine similarity scores from a surrogate FR model to maximally degrade identity recognition. We demonstrate that with approximately 10,000 queries to a black-box ArcFace model, the resulting GaP achieves a high attack success rate in both digital and real-world physical tests. Critically, the attack shows strong transferability, successfully deceiving an entirely unseen FaceNet model. Our work highlights a practical and severe vulnerability, proving that robust, transferable attacks can be crafted with limited knowledge of the target system.
Abstract (translated)
针对面部识别（FR）系统的对抗性攻击构成了重要的安全威胁，但大多数此类攻击局限于数字领域或需要白盒访问权限。我们引入了一种名为GaP（高斯补丁，Gaussian Patch）的新方法，在严格的黑盒设置下生成一种通用且物理可转移的对抗性补丁。我们的方法采用查询效率高的零阶贪婪算法，迭代构造用于额头区域的对称灰度图案。通过逐步添加高斯斑点，并仅依据替代面部识别模型提供的余弦相似度分数来优化该图案，以最大限度地降低身份识别准确性。我们展示了在大约10,000次针对黑盒ArcFace模型的查询后生成的GaP补丁，在数字和现实世界物理测试中均实现了较高的攻击成功率。重要的是，这种攻击展现出强大的可转移性，并成功欺骗了一个完全未见过的FaceNet模型。我们的工作突显了一种实际且严重的漏洞，证明了即使对目标系统的了解有限，也可以设计出稳健、可转移的对抗攻击。
URL
https://arxiv.org/abs/2509.23198
PDF
-
ExpFace: Exponential Angular Margin Loss for Deep Face Recognition
2025-09-24 04:08:19Jinhui Zheng, Xueyuan GongAbstract
Face recognition is an open-set problem requiring high discriminative power to ensure that intra-class distances remain smaller than inter-class distances. Margin-based softmax losses, such as SphereFace, CosFace, and ArcFace, have been widely adopted to enhance intra-class compactness and inter-class separability, yet they overlook the impact of noisy samples. By examining the distribution of samples in the angular space, we observe that clean samples predominantly cluster in the center region, whereas noisy samples tend to shift toward the peripheral region. Motivated by this observation, we propose the Exponential Angular Margin Loss (ExpFace), which introduces an angular exponential term as the margin. This design applies a larger penalty in the center region and a smaller penalty in the peripheral region within the angular space, thereby emphasizing clean samples while suppressing noisy samples. We present a unified analysis of ExpFace and classical margin-based softmax losses in terms of margin embedding forms, similarity curves, and gradient curves, showing that ExpFace not only avoids the training instability of SphereFace and the non-monotonicity of ArcFace, but also exhibits a similarity curve that applies penalties in the same manner as the decision boundary in the angular space. Extensive experiments demonstrate that ExpFace achieves state-of-the-art performance. To facilitate future research, we have released the source code at: this https URL.
Abstract (translated)
人脸识别是一个开放集问题，需要强大的判别能力以确保同类样本之间的距离小于不同类样本之间的距离。为了提高这一性能，基于余弦softmax损失的方法（如SphereFace、CosFace和ArcFace）已经被广泛采用来增强同类紧凑性和异类可分离性。然而，这些方法在处理噪声样本时表现不佳。 通过对角空间中样本分布的分析发现，干净样本主要聚集在中心区域，而噪声样本则倾向于移向边缘区域。基于这一观察，我们提出了一种指数角度余量损失（ExpFace），它引入了一个角度指数项作为余量。这种设计在角空间中的中心区域施加更大的惩罚，在边缘区域施加较小的惩罚，从而强调干净样本的同时抑制噪声样本。 我们在余量嵌入形式、相似性曲线和梯度曲线上对ExpFace与传统基于边际的softmax损失进行了统一分析。结果显示，ExpFace不仅避免了SphereFace训练时的不稳定性和ArcFace非单调性的缺点，还展示出一种与其角空间决策边界施加惩罚方式一致的相似性曲线。 广泛的实验表明，ExpFace在性能上达到了最先进的水平。为了促进未来的研究工作，我们已经发布了源代码：[请参考原文献或该领域的研究论文获取准确链接]。
URL
https://arxiv.org/abs/2509.19753
PDF
-
FROQ: Observing Face Recognition Models for Efficient Quality Assessment
2025-09-22 12:29:44\v{Z}iga Babnik, Deepak Kumar Jain, Peter Peer, Vitomir \v{S}trucAbstract
Face Recognition (FR) plays a crucial role in many critical (high-stakes) applications, where errors in the recognition process can lead to serious consequences. Face Image Quality Assessment (FIQA) techniques enhance FR systems by providing quality estimates of face samples, enabling the systems to discard samples that are unsuitable for reliable recognition or lead to low-confidence recognition decisions. Most state-of-the-art FIQA techniques rely on extensive supervised training to achieve accurate quality estimation. In contrast, unsupervised techniques eliminate the need for additional training but tend to be slower and typically exhibit lower performance. In this paper, we introduce FROQ (Face Recognition Observer of Quality), a semi-supervised, training-free approach that leverages specific intermediate representations within a given FR model to estimate face-image quality, and combines the efficiency of supervised FIQA models with the training-free approach of unsupervised methods. A simple calibration step based on pseudo-quality labels allows FROQ to uncover specific representations, useful for quality assessment, in any modern FR model. To generate these pseudo-labels, we propose a novel unsupervised FIQA technique based on sample perturbations. Comprehensive experiments with four state-of-the-art FR models and eight benchmark datasets show that FROQ leads to highly competitive results compared to the state-of-the-art, achieving both strong performance and efficient runtime, without requiring explicit training.
Abstract (translated)
人脸识别（FR）在许多关键应用中扮演着重要角色，识别过程中的错误可能导致严重后果。人脸图像质量评估（FIQA）技术通过提供面部样本的质量估计来增强人脸识别系统，使得系统能够丢弃不适合可靠识别或导致低置信度识别决策的样本。 大多数最新的FIQA技术依赖于广泛的监督训练以实现准确的质量估算。相比之下，无监督技术消除了对额外训练的需求，但往往运行较慢且通常表现出较低性能。在本文中，我们介绍了一种半监督、无需训练的方法FROQ（Face Recognition Observer of Quality），该方法利用给定人脸识别模型中的特定中间表示来估计人脸图像的质量，并结合了基于监督FIQA模型的效率和无监督方法的免训特性。 通过基于伪质量标签的简单校准步骤，FROQ能够揭示任何现代人脸识别模型中对质量评估有用的特定表示。为了生成这些伪标签，我们提出了一种基于样本扰动的新颖无监督FIQA技术。在四个最先进的FR模型和八个基准数据集上进行的全面实验表明，与现有方法相比，FROQ可以实现高度竞争的结果，在不需显式训练的情况下同时提供强大的性能和高效的运行时间。
URL
https://arxiv.org/abs/2509.17689
PDF
-
Trade-offs in Cross-Domain Generalization of Foundation Model Fine-Tuned for Biometric Applications
2025-09-18 12:58:18Tahar Chettaoui, Naser Damer, Fadi BoutrosAbstract
Foundation models such as CLIP have demonstrated exceptional zero- and few-shot transfer capabilities across diverse vision tasks. However, when fine-tuned for highly specialized biometric tasks, face recognition (FR), morphing attack detection (MAD), and presentation attack detection (PAD), these models may suffer from over-specialization. Thus, they may lose one of their foundational strengths, cross-domain generalization. In this work, we systematically quantify these trade-offs by evaluating three instances of CLIP fine-tuned for FR, MAD, and PAD. We evaluate each adapted model as well as the original CLIP baseline on 14 general vision datasets under zero-shot and linear-probe protocols, alongside common FR, MAD, and PAD benchmarks. Our results indicate that fine-tuned models suffer from over-specialization, especially when fine-tuned for complex tasks of FR. Also, our results pointed out that task complexity and classification head design, multi-class (FR) vs. binary (MAD and PAD), correlate with the degree of catastrophic forgetting. The FRoundation model with the ViT-L backbone outperforms other approaches on the large-scale FR benchmark IJB-C, achieving an improvement of up to 58.52%. However, it experiences a substantial performance drop on ImageNetV2, reaching only 51.63% compared to 69.84% achieved by the baseline CLIP model. Moreover, the larger CLIP architecture consistently preserves more of the model's original generalization ability than the smaller variant, indicating that increased model capacity may help mitigate over-specialization.
Abstract (translated)
基础模型（如CLIP）在各种视觉任务中展示了卓越的零样本和少量样本迁移能力。然而，当这些模型被微调用于高度专业化的生物识别任务（例如面部识别、变形攻击检测和呈现攻击检测）时，它们可能会出现过度专业化的问题，从而丧失其跨领域泛化的基本优势。 在这项工作中，我们系统地量化了这种权衡，通过评估三组CLIP模型在经过面部识别、形态攻击检测和呈现攻击检测微调后的性能。我们在14个通用视觉数据集以及常见的面部识别、变形攻击检测和呈现攻击检测基准上，使用零样本和线性探测协议对每个适应的模型及原始CLIP基线进行了评估。 我们的结果显示，经过微调的模型会遭受过度专业化的问题，尤其是当它们用于复杂的面部识别任务时。此外，我们的研究还表明，任务复杂性和分类头的设计（多类 vs 二元）与灾难性遗忘的程度相关联。带有ViT-L骨干网络的基础模型在大规模面部识别基准IJB-C上表现优于其他方法，改进幅度高达58.52%。然而，在ImageNetV2数据集上，它的性能显著下降，仅达到51.63%，相比之下，原始的CLIP基线模型则达到了69.84%。 此外，较大的CLIP架构相比较小的变体在一定程度上更好地保持了模型原有的泛化能力，这表明增加模型容量可能有助于减轻过度专业化的问题。
URL
https://arxiv.org/abs/2509.14921
PDF
-
Sy-FAR: Symmetry-based Fair Adversarial Robustness
2025-09-16 10:39:42Haneen Najjar, Eyal Ronen, Mahmood SharifAbstract
Security-critical machine-learning (ML) systems, such as face-recognition systems, are susceptible to adversarial examples, including real-world physically realizable attacks. Various means to boost ML's adversarial robustness have been proposed; however, they typically induce unfair robustness: It is often easier to attack from certain classes or groups than from others. Several techniques have been developed to improve adversarial robustness while seeking perfect fairness between classes. Yet, prior work has focused on settings where security and fairness are less critical. Our insight is that achieving perfect parity in realistic fairness-critical tasks, such as face recognition, is often infeasible -- some classes may be highly similar, leading to more misclassifications between them. Instead, we suggest that seeking symmetry -- i.e., attacks from class $i$ to $j$ would be as successful as from $j$ to $i$ -- is more tractable. Intuitively, symmetry is a desirable because class resemblance is a symmetric relation in most domains. Additionally, as we prove theoretically, symmetry between individuals induces symmetry between any set of sub-groups, in contrast to other fairness notions where group-fairness is often elusive. We develop Sy-FAR, a technique to encourage symmetry while also optimizing adversarial robustness and extensively evaluate it using five datasets, with three model architectures, including against targeted and untargeted realistic attacks. The results show Sy-FAR significantly improves fair adversarial robustness compared to state-of-the-art methods. Moreover, we find that Sy-FAR is faster and more consistent across runs. Notably, Sy-FAR also ameliorates another type of unfairness we discover in this work -- target classes that adversarial examples are likely to be classified into become significantly less vulnerable after inducing symmetry.
Abstract (translated)
关键安全的机器学习（ML）系统，如面部识别系统，容易受到对抗性样本的影响，包括现实世界中可实现的实际攻击。已经提出了各种方法来增强ML系统的对抗鲁棒性；然而，它们通常会产生不公平的稳健性：从某些类或群体发起攻击往往比从其他类或群体更容易。为提高对抗鲁棒性同时寻求不同类别间的完美公平性，开发了几种技术。然而，先前的研究主要集中在安全和公平性相对不重要的场景中。 我们的见解是，在如面部识别等现实中的关键任务上实现完美的公平性往往是不可行的——某些类可能非常相似，导致它们之间更容易发生误分类。相反，我们建议寻求对称性——即从类别i到j发起的攻击与从类别j到i发起的成功程度相同——是更可行的目标。直观地说，对称性是可取的，因为在大多数领域中，类别之间的类似性是对称的关系。 此外，我们理论证明，在个体之间诱导出的对称性会导致任何子组集合之间的对称性，这与公平性的其他概念不同，后者往往难以实现群体间的公平性。为此，我们开发了一种名为Sy-FAR的技术，旨在促进这种对称性的同时优化对抗鲁棒性，并通过使用五个数据集和三种模型架构（包括针对目标攻击和非目标实际攻击）进行了广泛的评估。 实验结果显示，与现有的最先进的方法相比，Sy-FAR显著提高了公平的对抗鲁棒性。此外，我们发现Sy-FAR运行速度更快且结果更一致。值得注意的是，Sy-FAR还改善了我们在工作中发现的一种不公平现象——被分类为某些目标类别的对抗样本，在诱导对称性后变得明显不那么脆弱。
URL
https://arxiv.org/abs/2509.12939
PDF
-
Character-Centric Understanding of Animated Movies
2025-09-15 17:59:51Zhongrui Gui, Junyu Xie, Tengda Han, Weidi Xie, Andrew ZissermanAbstract
Animated movies are captivating for their unique character designs and imaginative storytelling, yet they pose significant challenges for existing recognition systems. Unlike the consistent visual patterns detected by conventional face recognition methods, animated characters exhibit extreme diversity in their appearance, motion, and deformation. In this work, we propose an audio-visual pipeline to enable automatic and robust animated character recognition, and thereby enhance character-centric understanding of animated movies. Central to our approach is the automatic construction of an audio-visual character bank from online sources. This bank contains both visual exemplars and voice (audio) samples for each character, enabling subsequent multi-modal character recognition despite long-tailed appearance distributions. Building on accurate character recognition, we explore two downstream applications: Audio Description (AD) generation for visually impaired audiences, and character-aware subtitling for the hearing impaired. To support research in this domain, we introduce CMD-AM, a new dataset of 75 animated movies with comprehensive annotations. Our character-centric pipeline demonstrates significant improvements in both accessibility and narrative comprehension for animated content over prior face-detection-based approaches. For the code and dataset, visit this https URL.
Abstract (translated)
动画电影因其独特的角色设计和富有想象力的叙事而引人入胜，但现有的识别系统在处理这些内容时面临重大挑战。与传统面部识别方法检测到的一致性视觉模式不同，动画角色在外貌、动作及变形方面展现出了极大的多样性。为此，我们提出了一种音频-视觉管道来实现自动且稳健的动画人物识别，并借此增强对动画电影中以角色为中心的理解。 我们的核心方法是通过在线资源自动构建一个集视觉范例和声音（音频）样本于一体的音频-视觉角色库。这使得即便是在出现长尾分布的情况下，也能够进行多模态的角色识别。在此基础上，我们探讨了两个下游应用：为视障观众生成音轨描述(AD)以及针对听障观众的具有角色意识的字幕。 为了支持该领域的研究，我们引入了一个名为CMD-AM的新数据集，其中包括75部动画电影及其全面注释。我们的以角色为中心的管道在提高无障碍访问和叙事理解方面相对于传统的基于面部检测的方法展示出了显著的进步。 有关代码和数据集，请访问提供的URL链接。
URL
https://arxiv.org/abs/2509.12204
PDF
-
A Lightweight Ensemble-Based Face Image Quality Assessment Method with Correlation-Aware Loss
2025-09-12 10:13:38MohammadAli Hamidi, Hadi Amirpour, Luigi Atzori, Christian TimmererAbstract
Face image quality assessment (FIQA) plays a critical role in face recognition and verification systems, especially in uncontrolled, real-world environments. Although several methods have been proposed, general-purpose no-reference image quality assessment techniques often fail to capture face-specific degradations. Meanwhile, state-of-the-art FIQA models tend to be computationally intensive, limiting their practical applicability. We propose a lightweight and efficient method for FIQA, designed for the perceptual evaluation of face images in the wild. Our approach integrates an ensemble of two compact convolutional neural networks, MobileNetV3-Small and ShuffleNetV2, with prediction-level fusion via simple averaging. To enhance alignment with human perceptual judgments, we employ a correlation-aware loss (MSECorrLoss), combining mean squared error (MSE) with a Pearson correlation regularizer. Our method achieves a strong balance between accuracy and computational cost, making it suitable for real-world deployment. Experiments on the VQualA FIQA benchmark demonstrate that our model achieves a Spearman rank correlation coefficient (SRCC) of 0.9829 and a Pearson linear correlation coefficient (PLCC) of 0.9894, remaining within competition efficiency constraints.
Abstract (translated)
人脸图像质量评估（FIQA）在人脸识别和验证系统中扮演着关键角色，特别是在不受控的现实世界环境中。尽管已经提出了几种方法，但通用的无参考图像质量评估技术往往无法捕捉到特定于人脸的退化情况。与此同时，最先进的FIQA模型通常计算密集度高，限制了其实用性。我们提出了一种轻量且高效的FIQA方法，旨在对野外的人脸图像进行感知评价。我们的方法结合了一个由两个紧凑型卷积神经网络（MobileNetV3-Small和ShuffleNetV2）组成的集成，并通过简单的平均在预测层面融合。为了增强与人类感知判断的一致性，我们采用了一种相关感知损失（MSECorrLoss），该损失将均方误差（MSE）与皮尔逊相关正则化相结合。我们的方法实现了准确性和计算成本之间的强大平衡，使其适合于实际部署。在VQualA FIQA基准测试上的实验表明，我们的模型达到了斯皮尔曼等级相关系数（SRCC）0.9829和皮尔森线性相关系数（PLCC）0.9894的性能水平，并且仍然符合竞赛效率约束条件。
URL
https://arxiv.org/abs/2509.10114
PDF
-
Computational Imaging for Enhanced Computer Vision
2025-09-10 16:02:42Humera Shaikh, Kaur JashanpreetAbstract
This paper presents a comprehensive survey of computational imaging (CI) techniques and their transformative impact on computer vision (CV) applications. Conventional imaging methods often fail to deliver high-fidelity visual data in challenging conditions, such as low light, motion blur, or high dynamic range scenes, thereby limiting the performance of state-of-the-art CV systems. Computational imaging techniques, including light field imaging, high dynamic range (HDR) imaging, deblurring, high-speed imaging, and glare mitigation, address these limitations by enhancing image acquisition and reconstruc- tion processes. This survey systematically explores the synergies between CI techniques and core CV tasks, including object detection, depth estimation, optical flow, face recognition, and keypoint detection. By analyzing the relationships between CI methods and their practical contributions to CV applications, this work highlights emerging opportunities, challenges, and future research directions. We emphasize the potential for task-specific, adaptive imaging pipelines that improve robustness, accuracy, and efficiency in real-world scenarios, such as autonomous navigation, surveillance, augmented reality, and robotics.
Abstract (translated)
本文全面回顾了计算成像（CI）技术及其对计算机视觉（CV）应用的变革性影响。传统成像方法在诸如低光、运动模糊或高动态范围场景等挑战条件下，往往无法提供高质量的视觉数据，从而限制了当前最先进的CV系统的表现能力。计算成像技术，包括光场成像、高动态范围（HDR）成像、去模糊处理、高速成像和消除眩光等方法，在增强图像采集和重建过程中解决了这些问题。 本文系统地探讨了CI技术和核心CV任务之间的协同效应，这些任务包括物体检测、深度估计、光学流分析、人脸识别以及关键点检测。通过分析CI方法与其在实际CV应用中的贡献关系，本研究强调了新兴的机会、挑战及未来的研究方向。我们特别关注特定任务的自适应成像管道的潜力，这将提高真实场景（如自主导航、监控、增强现实和机器人技术）中系统的鲁棒性、准确性和效率。
URL
https://arxiv.org/abs/2509.08712
PDF
-
Backdoor Poisoning Attack Against Face Spoofing Attack Detection Methods
2025-09-03 08:05:36Shota Iwamatsu, Koichi Ito, Takafumi AokiAbstract
Face recognition systems are robust against environmental changes and noise, and thus may be vulnerable to illegal authentication attempts using user face photos, such as spoofing attacks. To prevent such spoofing attacks, it is crucial to discriminate whether the input image is a live user image or a spoofed image prior to the face recognition process. Most existing spoofing attack detection methods utilize deep learning, which necessitates a substantial amount of training data. Consequently, if malicious data is injected into a portion of the training dataset, a specific spoofing attack may be erroneously classified as live, leading to false this http URL this paper, we propose a novel backdoor poisoning attack method to demonstrate the latent threat of backdoor poisoning within face anti-spoofing detection. The proposed method enables certain spoofing attacks to bypass detection by embedding features extracted from the spoofing attack's face image into a live face image without inducing any perceptible visual this http URL experiments conducted on public datasets, we demonstrate that the proposed method constitutes a realistic threat to existing spoofing attack detection systems.
Abstract (translated)
面部识别系统在面对环境变化和噪音时具有较强的鲁棒性，但这使得它们可能容易受到使用用户面部照片进行的非法认证尝试（如伪装攻击）的影响。为了防止这类伪装攻击，在执行面部识别之前必须区分输入图像是否为真实用户的活体图像或伪造图像。现有的大多数防伪检测方法都依赖于深度学习技术，这需要大量的训练数据。因此，如果恶意数据被注入到部分训练集中，可能会导致特定的伪装攻击错误地被分类为活体，从而引发误报。 本文提出了一种新颖的后门投毒攻击方法，展示了面部防伪检测中潜在的后门投毒威胁。该方法通过将来自伪造图像的特征嵌入到真实用户的活体图像中而不引起任何可察觉的视觉变化，使某些伪装攻击能够绕过检测。实验结果表明，在公共数据集上使用这种方法会对现有的防伪攻击检测系统构成切实的威胁。
URL
https://arxiv.org/abs/2509.03108
PDF
-
CryptoFace: End-to-End Encrypted Face Recognition
2025-08-30 03:11:54Wei Ao, Vishnu Naresh BoddetiAbstract
Face recognition is central to many authentication, security, and personalized applications. Yet, it suffers from significant privacy risks, particularly arising from unauthorized access to sensitive biometric data. This paper introduces CryptoFace, the first end-to-end encrypted face recognition system with fully homomorphic encryption (FHE). It enables secure processing of facial data across all stages of a face-recognition process--feature extraction, storage, and matching--without exposing raw images or features. We introduce a mixture of shallow patch convolutional networks to support higher-dimensional tensors via patch-based processing while reducing the multiplicative depth and, thus, inference latency. Parallel FHE evaluation of these networks ensures near-resolution-independent latency. On standard face recognition benchmarks, CryptoFace significantly accelerates inference and increases verification accuracy compared to the state-of-the-art FHE neural networks adapted for face recognition. CryptoFace will facilitate secure face recognition systems requiring robust and provable security. The code is available at this https URL.
Abstract (translated)
面部识别是许多认证、安全和个人化应用的核心，但它面临着严重的隐私风险，尤其是由于未经授权访问敏感的生物特征数据。本文介绍了CryptoFace，这是第一个使用全同态加密（FHE）实现端到端加密的脸部识别系统。它能够在脸部识别过程的所有阶段——从特征提取、存储到匹配——对面部数据进行安全处理，而不泄露原始图像或特征。 我们引入了一种浅层补丁卷积网络的混合方式，通过基于补丁的处理来支持更高维度的数据张量，并减少了乘法深度，从而降低了推理延迟。并行化FHE评估这些网络确保了接近分辨率无关性的延迟时间。在标准面部识别基准测试中，CryptoFace与目前适应脸部识别的最佳FHE神经网络相比，在加速推理和提高验证准确性方面取得了显著进步。 CryptoFace将有助于建立安全的面部识别系统，要求具备强大的和可证明的安全性保障。相关代码可以在提供的链接中获取。
URL
https://arxiv.org/abs/2509.00332
PDF
-
DemoBias: An Empirical Study to Trace Demographic Biases in Vision Foundation Models
2025-08-25 18:02:49Abu Sufian, Anirudha Ghosh, Debaditya Barman, Marco Leo, Cosimo DistanteAbstract
Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities across various downstream tasks, including biometric face recognition (FR) with description. However, demographic biases remain a critical concern in FR, as these foundation models often fail to perform equitably across diverse demographic groups, considering ethnicity/race, gender, and age. Therefore, through our work DemoBias, we conduct an empirical evaluation to investigate the extent of demographic biases in LVLMs for biometric FR with textual token generation tasks. We fine-tuned and evaluated three widely used pre-trained LVLMs: LLaVA, BLIP-2, and PaliGemma on our own generated demographic-balanced dataset. We utilize several evaluation metrics, like group-specific BERTScores and the Fairness Discrepancy Rate, to quantify and trace the performance disparities. The experimental results deliver compelling insights into the fairness and reliability of LVLMs across diverse demographic groups. Our empirical study uncovered demographic biases in LVLMs, with PaliGemma and LLaVA exhibiting higher disparities for Hispanic/Latino, Caucasian, and South Asian groups, whereas BLIP-2 demonstrated comparably consistent. Repository: this https URL.
Abstract (translated)
大型视觉语言模型（LVLM）在包括带描述的生物识别面部识别（FR）在内的各种下游任务中表现出色。然而，人口统计学偏见仍然是面部识别中的一个关键问题，因为这些基础模型往往无法在考虑种族/民族、性别和年龄等多样化的群体时实现公平性。因此，在我们的工作DemoBias中，我们通过实证评估来调查LVLM在生物特征FR任务中文本标记生成方面的种族偏见程度。我们在自动生成的均衡人口统计学数据集上对三个广泛使用的预训练LVLM进行了微调和评估：LLaVA、BLIP-2和PaliGemma。我们使用了多种评估指标，如特定群体的BERT得分和公平性差异率，以量化和追踪性能差距。实验结果为LVLM在不同人口统计学群体中的公平性和可靠性提供了深刻的见解。我们的实证研究发现，在LLaVA和PaliGemma中，西班牙裔/拉丁裔、高加索人种和南亚人群体之间的偏见程度较高，而BLIP-2则表现出相对一致的表现。 相关代码库链接：[此URL](this https URL)
URL
https://arxiv.org/abs/2508.19298
PDF
-
Paired-Sampling Contrastive Framework for Joint Physical-Digital Face Attack Detection
2025-08-20 18:05:49Andrei Balykin, Anvar Ganiev, Denis Kondranin, Kirill Polevoda, Nikolai Liudkevich, Artem PetrovAbstract
Modern face recognition systems remain vulnerable to spoofing attempts, including both physical presentation attacks and digital forgeries. Traditionally, these two attack vectors have been handled by separate models, each targeting its own artifacts and modalities. However, maintaining distinct detectors increases system complexity and inference latency and leaves systems exposed to combined attack vectors. We propose the Paired-Sampling Contrastive Framework, a unified training approach that leverages automatically matched pairs of genuine and attack selfies to learn modality-agnostic liveness cues. Evaluated on the 6th Face Anti-Spoofing Challenge Unified Physical-Digital Attack Detection benchmark, our method achieves an average classification error rate (ACER) of 2.10 percent, outperforming prior solutions. The framework is lightweight (4.46 GFLOPs) and trains in under one hour, making it practical for real-world deployment. Code and pretrained models are available at this https URL.
Abstract (translated)
现代面部识别系统仍然容易受到模拟攻击，包括物理展示攻击和数字伪造。传统上，这两种攻击向量分别由各自的模型处理，每个模型都针对特定的特征和模式。然而，维护独立的检测器会增加系统的复杂性和推理延迟，并使系统暴露于结合了两种攻击向量的风险之下。我们提出了配对采样对比框架，这是一种统一的训练方法，利用自动匹配的真实自拍照与模拟攻击图像对来学习不受模态限制的生命体征提示符。在第六届面部防伪造挑战赛（Unified Physical-Digital Attack Detection）基准测试中，我们的方法实现了2.10%的平均分类错误率(ACER)，优于以前的方法。该框架轻量级（4.46 GFLOPs），训练时间不到一小时，在实际部署中具有实用性。代码和预训练模型可在此链接获取：[原链接中的URL]
URL
https://arxiv.org/abs/2508.14980
PDF
-
ViT-FIQA: Assessing Face Image Quality using Vision Transformers
2025-08-19 15:50:07Andrea Atzori, Fadi Boutros, Naser DamerAbstract
Face Image Quality Assessment (FIQA) aims to predict the utility of a face image for face recognition (FR) systems. State-of-the-art FIQA methods mainly rely on convolutional neural networks (CNNs), leaving the potential of Vision Transformer (ViT) architectures underexplored. This work proposes ViT-FIQA, a novel approach that extends standard ViT backbones, originally optimized for FR, through a learnable quality token designed to predict a scalar utility score for any given face image. The learnable quality token is concatenated with the standard image patch tokens, and the whole sequence is processed via global self-attention by the ViT encoders to aggregate contextual information across all patches. At the output of the backbone, ViT-FIQA branches into two heads: (1) the patch tokens are passed through a fully connected layer to learn discriminative face representations via a margin-penalty softmax loss, and (2) the quality token is fed into a regression head to learn to predict the face sample's utility. Extensive experiments on challenging benchmarks and several FR models, including both CNN- and ViT-based architectures, demonstrate that ViT-FIQA consistently achieves top-tier performance. These results underscore the effectiveness of transformer-based architectures in modeling face image utility and highlight the potential of ViTs as a scalable foundation for future FIQA research this https URL.
Abstract (translated)
面部图像质量评估（FIQA）旨在预测给定的人脸图像对于人脸识别系统（FR）的实用性。当前最先进的FIQA方法主要依赖于卷积神经网络（CNN），而视觉变压器（ViT）架构的潜力尚未被充分探索。这项工作提出了一种新的方法——ViT-FIQA，它通过扩展标准的专为面部识别优化的ViT骨干网来实现这一目标，并引入了一个可学习的质量标记(token)，用于预测任意给定人脸图像的标量实用性分数。 该可学习的质量标记与标准图像块标记(concatenated)结合在一起，整个序列通过ViT编码器处理，使用全局自我注意机制来汇集所有块之间的上下文信息。在骨干网的输出处，ViT-FIQA分支为两个头：（1）块标记通过全连接层传递，以学习具有边际惩罚softmax损失的人脸表示；（2）质量标记被馈送到回归头中，用于预测人脸样本的实用性。 针对多个挑战性的基准和多种人脸识别模型（包括CNN和ViT架构），进行了广泛的实验。这些实验证明了ViT-FIQA始终能达到顶级性能。该结果强调了基于变换器的架构在建模面部图像实用性方面的有效性，并突显出ViTs作为未来FIQA研究可扩展基础的巨大潜力。 原文链接：[这里插入链接]（请将这里的文本替换为实际的URL）
URL
https://arxiv.org/abs/2508.13957
PDF
-
A Comprehensive Re-Evaluation of Biometric Modality Properties in the Modern Era
2025-08-19 14:40:52Rouqaiah Al-Refai, Pankaja Priya Ramasamy, Ragini Ramesh, Patricia Arias-Cabarcos, Philipp Terh\"orstAbstract
The rapid advancement of authentication systems and their increasing reliance on biometrics for faster and more accurate user verification experience, highlight the critical need for a reliable framework to evaluate the suitability of biometric modalities for specific applications. Currently, the most widely known evaluation framework is a comparative table from 1998, which no longer adequately captures recent technological developments or emerging vulnerabilities in biometric systems. To address these challenges, this work revisits the evaluation of biometric modalities through an expert survey involving 24 biometric specialists. The findings indicate substantial shifts in property ratings across modalities. For example, face recognition, shows improved ratings due to technological progress, while fingerprint, shows decreased reliability because of emerging vulnerabilities and attacks. Further analysis of expert agreement levels across rated properties highlighted the consistency of the provided evaluations and ensured the reliability of the ratings. Finally, expert assessments are compared with dataset-level uncertainty across 55 biometric datasets, revealing strong alignment in most modalities and underscoring the importance of integrating empirical evidence with expert insight. Moreover, the identified expert disagreements reveal key open challenges and help guide future research toward resolving them.
Abstract (translated)
身份验证系统的快速发展及其对生物识别技术的日益依赖，以实现更快、更准确的用户验证体验，凸显了评估特定应用中生物识别模式适用性可靠框架的关键需求。目前最广为人知的评估框架是一个来自1998年的比较表格，它已不再充分反映最近的技术进步或新兴的安全漏洞。为解决这些挑战，这项工作通过一项涉及24位生物识别专家的调查，重新审视了对生物特征模态的评价。 研究结果表明，不同模态的属性评分发生了显著变化。例如，面部识别由于技术的进步而获得了更高的评分，而指纹识别则因新兴的安全漏洞和攻击而可靠性下降。进一步分析专家在评级特性上的共识水平，突显出所提供评估的一致性，并确保了评级的可靠性。 最后，通过将专家评估与55个生物特征数据集中的不确定性进行对比，发现大多数模态之间存在很强的一致性，强调了结合实证证据和专家见解的重要性。此外，所识别的专家分歧揭示了一些关键的研究挑战，并有助于指导未来研究解决这些问题的方向。
URL
https://arxiv.org/abs/2508.13874
PDF
-
Two-Factor Authentication Smart Entryway Using Modified LBPH Algorithm
2025-08-19 08:28:40Zakiah Ayop, Wan Mohamad Hariz Bin Wan Mohamad Rosdi, Looi Wei Hua, Syarulnaziah Anawar, Nur Fadzilah OthmanAbstract
Face mask detection has become increasingly important recently, particularly during the COVID-19 pandemic. Many face detection models have been developed in smart entryways using IoT. However, there is a lack of IoT development on face mask detection. This paper proposes a two-factor authentication system for smart entryway access control using facial recognition and passcode verification and an automation process to alert the owner and activate the surveillance system when a stranger is detected and controls the system remotely via Telegram on a Raspberry Pi platform. The system employs the Local Binary Patterns Histograms for the full face recognition algorithm and modified LBPH algorithm for occluded face detection. On average, the system achieved an Accuracy of approximately 70%, a Precision of approximately 80%, and a Recall of approximately 83.26% across all tested users. The results indicate that the system is capable of conducting face recognition and mask detection, automating the operation of the remote control to register users, locking or unlocking the door, and notifying the owner. The sample participants highly accept it for future use in the user acceptance test.
Abstract (translated)
最近，面部口罩检测变得越来越重要，特别是在COVID-19大流行期间。许多智能入口处的面部识别模型已经使用物联网技术开发出来，然而，在面部口罩检测方面的物联网开发却相对不足。本文提出了一种用于智能入口门禁控制的双因素认证系统，该系统结合了面部识别和密码验证，并在检测到陌生人时通过自动化过程向所有者发出警报并激活监控系统，同时还可以通过Telegram平台远程控制树莓派上的整个系统运行。 该系统采用局部二值模式直方图（LBPH）算法进行全脸识别，并且使用修改后的局部二值模式算法来检测被遮挡的面部。经过测试，在所有用户中，平均准确率约为70%，精确度约为80%，召回率为约83.26%。 结果显示，该系统能够执行面部识别和口罩检测、远程控制操作自动化（包括注册用户、锁定或解锁门以及通知所有者）。在用户体验接受性测试中，参与者对该系统的未来使用表示高度认可。
URL
https://arxiv.org/abs/2508.13617
PDF
-
Unified Knowledge Distillation Framework: Fine-Grained Alignment and Geometric Relationship Preservation for Deep Face Recognition
2025-08-15 10:20:29Durgesh Mishra, Rishabh UikeyAbstract
Knowledge Distillation is crucial for optimizing face recognition models for deployment in computationally limited settings, such as edge devices. Traditional KD methods, such as Raw L2 Feature Distillation or Feature Consistency loss, often fail to capture both fine-grained instance-level details and complex relational structures, leading to suboptimal performance. We propose a unified approach that integrates two novel loss functions, Instance-Level Embedding Distillation and Relation-Based Pairwise Similarity Distillation. Instance-Level Embedding Distillation focuses on aligning individual feature embeddings by leveraging a dynamic hard mining strategy, thereby enhancing learning from challenging examples. Relation-Based Pairwise Similarity Distillation captures relational information through pairwise similarity relationships, employing a memory bank mechanism and a sample mining strategy. This unified framework ensures both effective instance-level alignment and preservation of geometric relationships between samples, leading to a more comprehensive distillation process. Our unified framework outperforms state-of-the-art distillation methods across multiple benchmark face recognition datasets, as demonstrated by extensive experimental evaluations. Interestingly, when using strong teacher networks compared to the student, our unified KD enables the student to even surpass the teacher's accuracy.
Abstract (translated)
知识蒸馏对于优化面部识别模型以适应计算资源受限的环境（如边缘设备）至关重要。传统的知识蒸馏方法，例如原始L2特征蒸馏或特征一致性损失，在捕捉细粒度实例级细节和复杂关系结构方面常常表现不佳，从而导致性能次优。我们提出了一种统一的方法，该方法整合了两种新颖的损失函数：实例级嵌入蒸馏（Instance-Level Embedding Distillation）和基于关系的成对相似性蒸馏（Relation-Based Pairwise Similarity Distillation）。 实例级嵌入蒸馏通过利用动态硬挖掘策略来调整单个特征嵌入，从而增强了从挑战性样本中学习的能力。基于关系的成对相似性蒸馏则通过成对相似性的关系捕获结构化信息，并采用记忆库机制和样本挖掘策略进行优化。 这种统一框架确保了有效的实例级对齐以及样本之间几何关系的保留，进而实现了更加全面的知识蒸馏过程。广泛的实验评估表明，我们的统一框架在多个基准面部识别数据集上优于现有的知识蒸馏方法。有趣的是，当使用强大的教师网络时，学生模型通过这种方法甚至可以超越教师网络的表现准确性。
URL
https://arxiv.org/abs/2508.11376
PDF
