# The Do's and Don'ts for CNN-based Face Verification

arXiv: https://arxiv.org/abs/1705.07426 · HTML: https://mlsanity.com/paper/1705.07426/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 May 2017.

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

- [Poisoning Web-Scale Training Datasets is Practical](https://mlsanity.com/paper/2302.10149/) (2023)
- [PASS: Protected Attribute Suppression System for Mitigating Bias in Face Recognition](https://mlsanity.com/paper/2108.03764/) (2021)
- [MeInGame: Create a Game Character Face from a Single Portrait](https://mlsanity.com/paper/2102.02371/) (2021)
- [MixFairFace: Towards Ultimate Fairness via MixFair Adapter in Face Recognition](https://mlsanity.com/paper/2211.15181/) (2022)
- [A Synthesis-Based Approach for Thermal-to-Visible Face Verification](https://mlsanity.com/paper/2108.09558/) (2021)
- [Distill and De-bias: Mitigating Bias in Face Verification using Knowledge Distillation](https://mlsanity.com/paper/2112.09786/) (2021)
- [Recent Advances in Deep Learning Techniques for Face Recognition](https://mlsanity.com/paper/2103.10492/) (2021)
- [Twin identification over viewpoint change: A deep convolutional neural network surpasses humans](https://mlsanity.com/paper/2207.05316/) (2022)
- [A Brief Survey on Person Recognition at a Distance](https://mlsanity.com/paper/2212.08969/) (2022)
- [Meet-in-the-middle: Multi-scale upsampling and matching for cross-resolution face recognition](https://mlsanity.com/paper/2211.15225/) (2022)
- [The Influence of the Other-Race Effect on Susceptibility to Face Morphing Attacks](https://mlsanity.com/paper/2204.12591/) (2022)
- [Analyzing the Impact of Shape & Context on the Face Recognition Performance of Deep Networks](https://mlsanity.com/paper/2208.02991/) (2022)
- [Revisiting Face Recognition for Monozygotic Twins: The Celeb Twins Test Set](https://mlsanity.com/paper/2609.01141/) (2026)

## Abstract (arXiv)

While the research community appears to have developed a consensus on the methods of acquiring annotated data, design and training of CNNs, many questions still remain to be answered. In this paper, we explore the following questions that are critical to face recognition research: (i) Can we train on still images and expect the systems to work on videos? (ii) Are deeper datasets better than wider datasets? (iii) Does adding label noise lead to improvement in performance of deep networks? (iv) Is alignment needed for face recognition? We address these questions by training CNNs using CASIA-WebFace, UMDFaces, and a new video dataset and testing on YouTube- Faces, IJB-A and a disjoint portion of UMDFaces datasets. Our new data set, which will be made publicly available, has 22,075 videos and 3,735,476 human annotated frames extracted from them.