SphereFace Revived: Unifying Hyperspherical Face Recognition

  title={SphereFace Revived: Unifying Hyperspherical Face Recognition},
  author={Weiyang Liu and Yandong Wen and Bhiksha Raj and Rita Singh and Adrian Weller},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
This paper addresses the deep face recognition problem under an open-set protocol, where ideal face features are expected to have smaller maximal intra-class distance than minimal inter-class distance under a suitably chosen metric space. To this end, hyperspherical face recognition, as a promising line of research, has attracted increasing attention and gradually become a major focus in face recognition research. As one of the earliest works in hyperspherical face recognition, SphereFace… 

Unifying Margin-Based Softmax Losses in Face Recognition

A theoretical and experimental framework to study the effect of margin penalties on angular softmax losses, which have led to state-of-the-art performance in face recognition, and a new multiplicative margin which performs comparably to previously proposed additive margins when the model is trained to convergence.

Deep Metric Learning Using Negative Sampling Probability Annealing

A novel negative sampling solution based on dynamic policy switching, referred to as negative sampling probability annealing, which aims to exploit the positives of all approaches.

Continual Learning by Modeling Intra-Class Variation

This work examines memory-based continual learning and identifies that large variation in the representation space is crucial for avoiding catastrophic forgetting and proposes to diversify representations by using two types of perturbations: model-agnostic variation and model-based variation.

Just Noticeable Difference Modeling for Face Recognition System

This work develops a novel JND prediction model to directly infer JND images for the FR system and achieves higher accuracy of JND map prediction compared with the state-of-the-art JND models, and is capable of saving more bits while maintaining the performance of theFR system compared with VTM-15.0.

Exploring the Limits of Hard Example Mining for ID Document to Selfie Matching

A metric learning way rather than a classification-based way to train the network on IvS datasets, aiming to avoid tremendous pressure on GPU resource is adopted, and a Super Batch (S-Batch) is proposed by aggregating many traditional batches together.

Structural Causal 3D Reconstruction

This paper restricts the structure of latent space to capture a topological causal ordering of latent factors and demonstrates that the latent space structure indeed serves as an implicit regularization and introduces an inductive bias beneficial for reconstruction.

SphereFace2: Binary Classification is All You Need for Deep Face Recognition

This paper identifies the discrepancy between training and evaluation in the existing multi-class classification framework and discusses the potential limitations caused by the “competitive” nature of softmax normalization, and proposes a novel binary classification training framework, termed SphereFace2, which effectively bridges the gap betweenTraining and evaluation.



ArcFace: Additive Angular Margin Loss for Deep Face Recognition

This paper presents arguably the most extensive experimental evaluation against all recent state-of-the-art face recognition methods on ten face recognition benchmarks, and shows that ArcFace consistently outperforms the state of the art and can be easily implemented with negligible computational overhead.

VGGFace2: A Dataset for Recognising Faces across Pose and Age

A new large-scale face dataset named VGGFace2 is introduced, which contains 3.31 million images of 9131 subjects, with an average of 362.6 images for each subject, and the automated and manual filtering stages to ensure a high accuracy for the images of each identity are described.

Additive Margin Softmax for Face Verification

A conceptually simple and intuitive learning objective function, i.e., additive margin softmax, for face verification, which performs better when the evaluation criteria are designed for very low false alarm rate.

CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognition

This work proposes a novel Adaptive Curriculum Learning loss (CurricularFace) that embeds the idea of curriculum learning into the loss function to achieve a novel training strategy for deep face recognition, which mainly addresses easy samples in the early training stage and hard ones in the later stage.

IARPA Janus Benchmark-B Face Dataset

The IARPA Janus Benchmark-B (NIST IJB-B) dataset is introduced, a superset of IJB -A that represents operational use cases including access point identification, forensic quality media searches, surveillance video searches, and clustering.

NormFace: L2 Hypersphere Embedding for Face Verification

This work identifies and study four issues related to normalization through mathematical analysis, which yields understanding and helps with parameter settings, and proposes two strategies for training using normalized features.

MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition

A benchmark task to recognize one million celebrities from their face images, by using all the possibly collected face images of this individual on the web as training data, which could lead to one of the largest classification problems in computer vision.

SphereFace: Deep Hypersphere Embedding for Face Recognition

This paper proposes the angular softmax (A-Softmax) loss that enables convolutional neural networks (CNNs) to learn angularly discriminative features in deep face recognition (FR) problem under open-set protocol.

Circle Loss: A Unified Perspective of Pair Similarity Optimization

The Circle loss is demonstrated, which has a unified formula for two elemental deep feature learning paradigms, learning with class-level labels and pair-wise labels, and the superiority of the Circle loss on a variety ofDeep feature learning tasks.

CosFace: Large Margin Cosine Loss for Deep Face Recognition

  • H. WangYitong Wang Wei Liu
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • 2018
This paper reformulates the softmax loss as a cosine loss by L2 normalizing both features and weight vectors to remove radial variations, based on which acosine margin term is introduced to further maximize the decision margin in the angular space, and achieves minimum intra-class variance and maximum inter- class variance by virtue of normalization and cosine decision margin maximization.