Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

computer-vision

Face Identification

3260 papers • 126 benchmarks • 313 datasets

Face identification is the task of matching a given face image to one in an existing database of faces. It is the second part of face recognition (the first part being detection). It is a one-to-many mapping: you have to find an unknown person in a database to find who that person is.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in face-identification

Trend

Dataset

Best Model

Actions

MegaFace

MegaFace

Trillion Pairs Dataset

Trillion Pairs Dataset

DroneSURF

Libraries

i

Use these libraries to find face-identification models and implementations

FaceOnLive/Face-Recognition-SDK-And…

7 papers 215

Datasets

MS-Celeb-1M

MegaFace

IJB-A

IJB-B

DroneSURF

LTFT

Subtasks

No subtasks available.

Most implemented papers

FaceNet: A unified embedding for face recognition and clustering

James Philbin, Florian Schroff, Dmitry Kalenichenko•Wed Mar 11 2015

A system that directly learns a mapping from face images to a compact Euclidean space where distances directly correspond to a measure offace similarity, and achieves state-of-the-art face recognition performance using only 128-bytes perface.

14102

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

DroneSURF

IJB-A

IJB-A

IJB-B

IJB-B

serengil/deepface

6 papers 11,960

tomas-gajarsky/facetorch

4 papers 488

HSE-asavchenko/face-emotion-recogni…

3 papers 658

deepinsight/insightface

2 papers 22,967

PetFace

PetFace

0

ArcFace: Additive Angular Margin Loss for Deep Face Recognition

Jiankang Deng, J. Guo, S. Zafeiriou•Mon Jan 22 2018

One of the main challenges in feature learning using Deep Convolutional Neural Networks (DCNNs) for large-scale face recognition is the design of appropriate loss functions that can enhance the discriminative power. Centre loss penalises the distance between deep features and their corresponding class centres in the Euclidean space to achieve intra-class compactness. SphereFace assumes that the linear transformation matrix in the last fully connected layer can be used as a representation of the class centres in the angular space and therefore penalises the angles between deep features and their corresponding weights in a multiplicative way. Recently, a popular line of research is to incorporate margins in well-established loss functions in order to maximise face class separability. In this paper, we propose an Additive Angular Margin Loss (ArcFace) to obtain highly discriminative features for face recognition. The proposed ArcFace has a clear geometric interpretation due to its exact correspondence to geodesic distance on a hypersphere. We present arguably the most extensive experimental evaluation against all recent state-of-the-art face recognition methods on ten face recognition benchmarks which includes a new large-scale image database with trillions of pairs and a large-scale video dataset. We show that ArcFace consistently outperforms the state of the art and can be easily implemented with negligible computational overhead. To facilitate future research, the code has been made available.

7001 0

SphereFace: Deep Hypersphere Embedding for Face Recognition

Yandong Wen, B. Raj, Le Song, Zhiding Yu, Weiyang Liu, Ming Li•Tue Apr 25 2017

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.

2996 0

A Light CNN for Deep Face Representation With Noisy Labels

Zhenan Sun, R. He, T. Tan, Xiang Wu•Sun Nov 08 2015

Experimental results show that the proposed framework can utilize large-scale noisy data to learn a Light model that is efficient in computational costs and storage spaces and achieves state-of-the-art results on various face benchmarks without fine-tuning.

1133 0

Network In Network

Shuicheng Yan, Min Lin, Qiang Chen•Sun Dec 15 2013

With enhanced local modeling via the micro network, the proposed deep network structure NIN is able to utilize global average pooling over feature maps in the classification layer, which is easier to interpret and less prone to overfitting than traditional fully connected layers.

6599 0

CosFace: Large Margin Cosine Loss for Deep Face Recognition

Zhifeng Li, Yitong Wang, H. Wang, Zheng Zhou, Xing Ji, Dihong Gong, Jin Zhou•Sun Jan 28 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.

2818 0

DeepID3: Face Recognition with Very Deep Neural Networks

Xiaogang Wang, Xiaoou Tang, Yi Sun, Ding Liang•Mon Feb 02 2015

This paper proposes two very deep neural network architectures, referred to as DeepID3, for face recognition, rebuilt from stacked convolution and inception layers proposed in VGG net and GoogLeNet to make them suitable to face recognition.

964 0

Partial FC: Training 10 Million Identities on a Single Machine

Xiang An, Xuhan Zhu, Yanghua Xiao, Lan Wu, Ming Zhang, Yuan Gao, Bin Qin, Debing Zhang, Ying Fu•Sat Oct 10 2020

This work theoretically analyze the upper limit of model parallelism in face recognition and proposes a load-balanced sparse distributed classification training method, Partial FC, which is capable of using a machine with only 8 Nvidia Tesla V100 GPUs to implement training on a face recognition data set with up to 29 million IDs.

259 0

FacePoseNet: Making a Case for Landmark-Free Face Alignment

G. Medioni, R. Nevatia, I. Masi, Tal Hassner, A. Tran, Feng-Ju Chang•Wed Aug 23 2017

The FacePoseNet (FPN) is claimed to be a far faster and far more accurate face alignment method than using facial landmark detectors, and aligns faces at a small fraction of the computational cost of comparably accurate landmark detectors.

125 0

GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations

S. Javed, Mohamad Alansari, Oussama Abdul Hay, Abdulhadi Shoufan, Yahya Zweiri, N. Werghi•Sat Dec 31 2022

Evaluation of GhostFaceNets using various benchmarks reveals that these models offer superior performance while requiring a computational complexity significantly lower than that of State-Of-The-Art (SOTA) big convolutional neural network (CNN) models, which can require hundreds of millions of FLOPs.

96 0

Adding a benchmark result helps the community track progress.

Face Identification | State-of-the-Art