# CosFace: Large Margin Cosine Loss for Deep Face Recognition

arXiv: https://arxiv.org/abs/1801.09414 · HTML: https://mlsanity.com/paper/1801.09414/

## Key facts

- Cited by 971 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jan 2018.

## Built on by (most cited first, 30 of 971)

- [AdaFace: Quality Adaptive Margin for Face Recognition](https://mlsanity.com/paper/2204.00964/) (2022)
- [MagFace: A Universal Representation for Face Recognition and Quality Assessment](https://mlsanity.com/paper/2103.06627/) (2021)
- [Generalizing Face Forgery Detection with High-frequency Features](https://mlsanity.com/paper/2103.12376/) (2021)
- [Mitigating Neural Network Overconfidence with Logit Normalization](https://mlsanity.com/paper/2205.09310/) (2022)
- [SimSwap: An Efficient Framework For High Fidelity Face Swapping](https://mlsanity.com/paper/2106.06340/) (2021)
- [WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2103.04098/) (2021)
- [FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding](https://mlsanity.com/paper/2103.05950/) (2021)
- [Rethinking Visual Geo-localization for Large-Scale Applications](https://mlsanity.com/paper/2204.02287/) (2022)
- [ElasticFace: Elastic Margin Loss for Deep Face Recognition](https://mlsanity.com/paper/2109.09416/) (2021)
- [MegaPortraits: One-shot Megapixel Neural Head Avatars](https://mlsanity.com/paper/2207.07621/) (2022)
- [Frequency Domain Model Augmentation for Adversarial Attack](https://mlsanity.com/paper/2207.05382/) (2022)
- [How to Exploit Hyperspherical Embeddings for Out-of-Distribution Detection?](https://mlsanity.com/paper/2203.04450/) (2022)
- [Deformable ProtoPNet: An Interpretable Image Classifier Using Deformable Prototypes](https://mlsanity.com/paper/2111.15000/) (2021)
- [DCFace: Synthetic Face Generation with Dual Condition Diffusion Model](https://mlsanity.com/paper/2304.07060/) (2023)
- [EigenPlaces: Training Viewpoint Robust Models for Visual Place Recognition](https://mlsanity.com/paper/2308.10832/) (2023)
- [SynFace: Face Recognition with Synthetic Data](https://mlsanity.com/paper/2108.07960/) (2021)
- [Protecting Celebrities from DeepFake with Identity Consistency Transformer](https://mlsanity.com/paper/2203.01318/) (2022)
- [Neural Collapse: A Review on Modelling Principles and Generalization](https://mlsanity.com/paper/2206.04041/) (2022)
- [LowKey: Leveraging Adversarial Attacks to Protect Social Media Users from Facial Recognition](https://mlsanity.com/paper/2101.07922/) (2021)
- [Adversarial Sticker: A Stealthy Attack Method in the Physical World](https://mlsanity.com/paper/2104.06728/) (2021)
- [DigiFace-1M: 1 Million Digital Face Images for Face Recognition](https://mlsanity.com/paper/2210.02579/) (2022)
- [ReSSL: Relational Self-Supervised Learning with Weak Augmentation](https://mlsanity.com/paper/2107.09282/) (2021)
- [Few-Shot Class-Incremental Learning from an Open-Set Perspective](https://mlsanity.com/paper/2208.00147/) (2022)
- [Improving Transferability of Adversarial Patches on Face Recognition with Generative Models](https://mlsanity.com/paper/2106.15058/) (2021)
- [Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting](https://mlsanity.com/paper/2203.08354/) (2022)
- [Are All Losses Created Equal: A Neural Collapse Perspective](https://mlsanity.com/paper/2210.02192/) (2022)
- [DiffFace: Diffusion-based Face Swapping with Facial Guidance](https://mlsanity.com/paper/2212.13344/) (2022)
- [FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation](https://mlsanity.com/paper/2102.12867/) (2021)
- [Improving Vision Transformers by Revisiting High-frequency Components](https://mlsanity.com/paper/2204.00993/) (2022)
- [A Survey on Long-Tailed Visual Recognition](https://mlsanity.com/paper/2205.13775/) (2022)

## Abstract (arXiv)

Face recognition has made extraordinary progress owing to the advancement of deep convolutional neural networks (CNNs). The central task of face recognition, including face verification and identification, involves face feature discrimination. However, the traditional softmax loss of deep CNNs usually lacks the power of discrimination. To address this problem, recently several loss functions such as center loss, large margin softmax loss, and angular softmax loss have been proposed. All these improved losses share the same idea: maximizing inter-class variance and minimizing intra-class variance. In this paper, we propose a novel loss function, namely large margin cosine loss (LMCL), to realize this idea from a different perspective. More specifically, we reformulate the softmax loss as a cosine loss by L2 normalizing both features and weight vectors to remove radial variations, based on which a cosine margin term is introduced to further maximize the decision margin in the angular space. As a result, minimum intra-class variance and maximum inter-class variance are achieved by virtue of normalization and cosine decision margin maximization. We refer to our model trained with LMCL as CosFace. Extensive experimental evaluations are conducted on the most popular public-domain face recognition datasets such as MegaFace Challenge, Youtube Faces (YTF) and Labeled Face in the Wild (LFW). We achieve the state-of-the-art performance on these benchmarks, which confirms the effectiveness of our proposed approach.