Triplet Similarity Embedding for Face Verification

Abstract

In this work, we present an unconstrained face verification algorithm and evaluate it on the recently released IJB-A dataset that aims to push the boundaries of face verification methods. The proposed algorithm couples a deep CNN-based approach with a low-dimensional discriminative embedding learnt using triplet similarity constraints in a large margin fashion. Aside from yielding performance improvement, this embedding provides significant advantages in terms of memory and post-processing operations like hashing and visualization. Experiments on the IJB-A dataset show that the proposed algorithm outperforms state of the art methods in verification and identification metrics, while requiring less training time.

References in corpus (4)

• Learning Transferable Features with Deep Adaptation Networks
• Learning Face Representation from Scratch
• Face Search at Scale: 80 Million Gallery
• Unconstrained Face Verification using Deep CNN Features

Cited by in corpus (12)

• Deep Face Recognition: A Survey
• Face Recognition: From Traditional to Deep Learning Methods
• Do We Really Need to Collect Millions of Faces for Effective Face Recognition?
• von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification
• Racial Faces in-the-Wild: Reducing Racial Bias by Information Maximization Adaptation Network
• A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion
• Siamese Capsule Networks
• Ulixes: Facial Recognition Privacy with Adversarial Machine Learning
• SSDL: Self-Supervised Domain Learning for Improved Face Recognition
• Angular Learning: Toward Discriminative Embedded Features
• Face Identification and Clustering
• ClusterFace: Joint Clustering and Classification for Set-Based Face Recognition