Abstract:Feature extractors significantly impact the performance of biometric systems. In the field of hand gesture authentication, existing studies focus on improving the model architectures and behavioral characteristic representation methods to enhance their feature extractors. However, loss functions, which can guide extractors to produce more discriminative identity features, are neglected. In this paper, we improve the margin-based Softmax loss functions, which are mainly designed for face authentication, in two aspects to form a new loss function for hand gesture authentication. First, we propose to replace the commonly used cosine function in the margin-based Softmax losses with a linear function to measure the similarity between identity features and proxies (the weight matrix of Softmax, which can be viewed as class centers). With the linear function, the main gradient magnitude decreases monotonically as the quality of the model improves during training, thus allowing the model to be quickly optimized in the early stage and precisely fine-tuned in the late stage. Second, we design an adaptive margin scheme to assign margin penalties to different samples according to their separability and the model quality in each iteration. Our adaptive margin scheme constrains the gradient magnitude. It can reduce radical (excessively large) gradient magnitudes and provide moderate (not too small) gradient magnitudes for model optimization, contributing to more stable training. The linear function and the adaptive margin scheme are complementary. Combining them, we obtain the proposed linear adaptive additive angular margin (L3AM) loss. To demonstrate the effectiveness of L3AM loss, we conduct extensive experiments on seven hand-related authentication datasets, compare it with 25 state-of-the-art (SOTA) loss functions, and apply it to eight SOTA hand gesture authentication models. The experimental results show that L3AM loss further improves the performance of the eight authentication models and outperforms the 25 losses. The code is available at https://github.com/SCUT-BIP-Lab/L3AM.

Pose Adaptive Margin Loss for Face Verification

InterFace:Adjustable Angular Margin Inter-class Loss for Deep Face Recognition

Robust Face Recognition by Fusion Local Singular Value Feature and Deformable Model

Margin Loss: Making Faces More Separable

Angle-based Softmax Loss for Face Verification

HARD SAMPLES BASED MARGIN LOSS FOR FACE VERIFICATION d

Probabilistic Elastic Matching for Pose Variant Face Verification

Deep Representation for Partially Occluded Face Verification

Class-Variant Margin Normalized Softmax Loss for Deep Face Recognition

Deep Face Recognition Using Adaptively-Weighted Verification Loss Function.

Companion Guided Soft Margin for Face Recognition

Inter-class Angular Margin Loss for Face Recognition

X2-Softmax: Margin adaptive loss function for face recognition

PGM-face: Pose-guided Margin Loss for Cross-Pose Face Recognition

PoseFace: Pose-Invariant Features and Pose-Adaptive Loss for Face Recognition

Accurate and Reliable Facial Expression Recognition Using Advanced Softmax Loss with Fixed Weights

L3AM: Linear Adaptive Additive Angular Margin Loss for Video-Based Hand Gesture Authentication

Learning Towards The Largest Margins.

Frontal-Centers Guided Face: Boosting Face Recognition by Learning Pose-Invariant Features

Large Margin Softmax Loss for Speaker Verification

Makeup Removal for Face Verification Based Upon Contrastive Learning