# Crystal Loss and Quality Pooling for Unconstrained Face Verification and Recognition

arXiv: https://arxiv.org/abs/1804.01159 · HTML: https://mlsanity.com/paper/1804.01159/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Apr 2018.

## Built on by (most cited first, 13 of 13)

- [WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2103.04098/) (2021)
- [von Mises-Fisher Loss: An Exploration of Embedding Geometries for Supervised Learning](https://mlsanity.com/paper/2103.15718/) (2021)
- [KeyPoint Relative Position Encoding for Face Recognition](https://mlsanity.com/paper/2403.14852/) (2024)
- [QMagFace: Simple and Accurate Quality-Aware Face Recognition](https://mlsanity.com/paper/2111.13475/) (2021)
- [SphereFace Revived: Unifying Hyperspherical Face Recognition](https://mlsanity.com/paper/2109.05565/) (2021)
- [SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition](https://mlsanity.com/paper/2205.12010/) (2022)
- [EQFace: A Simple Explicit Quality Network for Face Recognition](https://mlsanity.com/paper/2105.00634/) (2021)
- [NPT-Loss: A Metric Loss with Implicit Mining for Face Recognition](https://mlsanity.com/paper/2103.03503/) (2021)
- [Twin identification over viewpoint change: A deep convolutional neural network surpasses humans](https://mlsanity.com/paper/2207.05316/) (2022)
- [Template-based Multi-Domain Face Recognition](https://mlsanity.com/paper/2409.09832/) (2024)
- [The Influence of the Other-Race Effect on Susceptibility to Face Morphing Attacks](https://mlsanity.com/paper/2204.12591/) (2022)
- [An Empirical Study on Clustering Pretrained Embeddings: Is Deep Strictly Better?](https://mlsanity.com/paper/2211.05183/) (2022)
- [Introducing the DOME Activation Functions](https://mlsanity.com/paper/2109.14798/) (2021)

## Abstract (arXiv)

In recent years, the performance of face verification and recognition systems based on deep convolutional neural networks (DCNNs) has significantly improved. A typical pipeline for face verification includes training a deep network for subject classification with softmax loss, using the penultimate layer output as the feature descriptor, and generating a cosine similarity score given a pair of face images or videos. The softmax loss function does not optimize the features to have higher similarity score for positive pairs and lower similarity score for negative pairs, which leads to a performance gap. In this paper, we propose a new loss function, called Crystal Loss, that restricts the features to lie on a hypersphere of a fixed radius. The loss can be easily implemented using existing deep learning frameworks. We show that integrating this simple step in the training pipeline significantly improves the performance of face verification and recognition systems. We achieve state-of-the-art performance for face verification and recognition on challenging LFW, IJB-A, IJB-B and IJB-C datasets over a large range of false alarm rates (10-1 to 10-7).