# Learning from Millions of 3D Scans for Large-scale 3D Face Recognition

arXiv: https://arxiv.org/abs/1711.05942 · HTML: https://mlsanity.com/paper/1711.05942/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Nov 2017.

## Built on by (most cited first, 22 of 22)

- [Advances in adversarial attacks and defenses in computer vision: A survey](https://mlsanity.com/paper/2108.00401/) (2021)
- [Physical-World Optical Adversarial Attacks on 3D Face Recognition](https://mlsanity.com/paper/2205.13412/) (2022)
- [Weakly-Supervised Multi-Face 3D Reconstruction](https://mlsanity.com/paper/2101.02000/) (2021)
- [Recent Advances in Deep Learning Techniques for Face Recognition](https://mlsanity.com/paper/2103.10492/) (2021)
- [High-Accuracy RGB-D Face Recognition via Segmentation-Aware Face Depth Estimation and Mask-Guided Attention Network](https://mlsanity.com/paper/2112.11713/) (2021)
- [A Survey on Physical Adversarial Attacks against Face Recognition Systems](https://mlsanity.com/paper/2410.16317/) (2024)
- [AFNet-M: Adaptive Fusion Network with Masks for 2D+3D Facial Expression Recognition](https://mlsanity.com/paper/2205.11785/) (2022)
- [Face-GCN: A Graph Convolutional Network for 3D Dynamic Face Identification/Recognition](https://mlsanity.com/paper/2104.09145/) (2021)
- [MFEViT: A Robust Lightweight Transformer-based Network for Multimodal 2D+3D Facial Expression Recognition](https://mlsanity.com/paper/2109.13086/) (2021)
- [Generating Unrestricted 3D Adversarial Point Clouds](https://mlsanity.com/paper/2111.08973/) (2021)
- [Generating 2D and 3D Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution](https://mlsanity.com/paper/2211.13964/) (2022)
- [3D Face Recognition: A Survey](https://mlsanity.com/paper/2108.11082/) (2021)
- [A Survey of Face Recognition](https://mlsanity.com/paper/2212.13038/) (2022)
- [DiFace: Cross-Modal Face Recognition through Controlled Diffusion](https://mlsanity.com/paper/2312.01367/) (2023)
- [Generative Landmarks Guided Eyeglasses Removal 3D Face Reconstruction](https://mlsanity.com/paper/2412.19848/) (2024)
- [DAP-MAE: Domain-Adaptive Point Cloud Masked Autoencoder for Effective Cross-Domain Learning](https://mlsanity.com/paper/2510.21635/) (2025)
- [Few-Data Guided Learning Upon End-to-End Point Cloud Network for 3D Face Recognition](https://mlsanity.com/paper/2103.16927/) (2021)
- [VA-GCN: A Vector Attention Graph Convolution Network for learning on Point Clouds](https://mlsanity.com/paper/2106.00227/) (2021)
- [Generating Dataset For Large-scale 3D Facial Emotion Recognition](https://mlsanity.com/paper/2109.08043/) (2021)
- [Confidence-Aware RGB-D Face Recognition via Virtual Depth Synthesis](https://mlsanity.com/paper/2403.06529/) (2024)
- [Face recognition on point cloud with cgan-top for denoising](https://mlsanity.com/paper/2506.06864/) (2025)
- [GFT-GCN: Privacy-Preserving 3D Face Mesh Recognition with Spectral Diffusion](https://mlsanity.com/paper/2511.19958/) (2025)

## Abstract (arXiv)

Deep networks trained on millions of facial images are believed to be closely approaching human-level performance in face recognition. However, open world face recognition still remains a challenge. Although, 3D face recognition has an inherent edge over its 2D counterpart, it has not benefited from the recent developments in deep learning due to the unavailability of large training as well as large test datasets. Recognition accuracies have already saturated on existing 3D face datasets due to their small gallery sizes. Unlike 2D photographs, 3D facial scans cannot be sourced from the web causing a bottleneck in the development of deep 3D face recognition networks and datasets. In this backdrop, we propose a method for generating a large corpus of labeled 3D face identities and their multiple instances for training and a protocol for merging the most challenging existing 3D datasets for testing. We also propose the first deep CNN model designed specifically for 3D face recognition and trained on 3.1 Million 3D facial scans of 100K identities. Our test dataset comprises 1,853 identities with a single 3D scan in the gallery and another 31K scans as probes, which is several orders of magnitude larger than existing ones. Without fine tuning on this dataset, our network already outperforms state of the art face recognition by over 10%. We fine tune our network on the gallery set to perform end-to-end large scale 3D face recognition which further improves accuracy. Finally, we show the efficacy of our method for the open world face recognition problem.