Search

Private voice communication often contains sensitive information, making it critical to ensure that only authorized users have access to such calls. Unfortunately, current authentication mechanisms, such as PIN-based passwords, fingerprint recognition, and face recognition, fail to authenticate the call receiver, leaving a gap in security. To fill the gap, we present EarPass, a secure and implicit call receiver authentication scheme designed for smartphones. EarPass sends inaudible acoustic signals through the earpiece speaker to actively sense the outer ear, and records echoes using the top microphone. It focuses on extracting ear-related signals from echoes and performs spectrogram analysis in the magnitude and phase domains. To overcome posture and position variability, EarPass utilizes a learning-based feature extractor for extracting representative features, and a one-class classifier for authentication. EarPass does not increase any burdens on users or change users' call answering habits. Furthermore, it does not require extra devices but only uses the speaker and microphone on the smartphone. We conducted comprehensive experiments to evaluate EarPass's effectiveness and security. Our results show that EarPass can achieve a balanced accuracy of 96.95% and an equal error rate of 1.53%. Additionally, EarPass exhibits resilience against potential attacks, including zero-effort attacks and mimicry attacks.

Synthetic data is gaining increasing relevance for training machine learning models. This is mainly motivated due to several factors such as the lack of real data and intra-class variability, time and errors produced in manual labeling, and in some cases privacy concerns, among others. This paper presents an overview of the 2nd edition of the Face Recognition Challenge in the Era of Synthetic Data (FRCSyn) organized at CVPR 2024. FRCSyn aims to investigate the use of synthetic data in face recognition to address current technological limitations, including data privacy concerns, demographic biases, generalization to novel scenarios, and performance constraints in challenging situations such as aging, pose variations, and occlusions. Unlike the 1st edition, in which synthetic data from DCFace and GANDiffFace methods was only allowed to train face recognition systems, in this 2nd edition we propose new sub-tasks that allow participants to explore novel face generative methods. The outcomes of the 2nd FRCSyn Challenge, along with the proposed experimental protocol and benchmarking contribute significantly to the application of synthetic data to face recognition., Comment: arXiv admin note: text overlap with arXiv:2311.10476

Transformers have emerged as the superior choice for face recognition tasks, but their insufficient platform acceleration hinders their application on mobile devices. In contrast, Convolutional Neural Networks (CNNs) capitalize on hardware-compatible acceleration libraries. Consequently, it has become indispensable to preserve the distillation efficacy when transferring knowledge from a Transformer-based teacher model to a CNN-based student model, known as Cross-Architecture Knowledge Distillation (CAKD). Despite its potential, the deployment of CAKD in face recognition encounters two challenges: 1) the teacher and student share disparate spatial information for each pixel, obstructing the alignment of feature space, and 2) the teacher network is not trained in the role of a teacher, lacking proficiency in handling distillation-specific knowledge. To surmount these two constraints, 1) we first introduce a Unified Receptive Fields Mapping module (URFM) that maps pixel features of the teacher and student into local features with unified receptive fields, thereby synchronizing the pixel-wise spatial information of teacher and student. Subsequently, 2) we develop an Adaptable Prompting Teacher network (APT) that integrates prompts into the teacher, enabling it to manage distillation-specific knowledge while preserving the model's discriminative capacity. Extensive experiments on popular face benchmarks and two large-scale verification sets demonstrate the superiority of our method.