# BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs

arXiv: https://arxiv.org/abs/1907.05047 · HTML: https://mlsanity.com/paper/1907.05047/

## Key facts

- Cited by 74 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Jul 2019.

## Built on by (most cited first, 30 of 74)

- [Deepfake Video Detection Using Convolutional Vision Transformer](https://mlsanity.com/paper/2102.11126/) (2021)
- [OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association](https://mlsanity.com/paper/2103.02440/) (2021)
- [Towards Solving the DeepFake Problem : An Analysis on Improving DeepFake Detection using Dynamic Face Augmentation](https://mlsanity.com/paper/2102.09603/) (2021)
- [Smart at what cost? Characterising Mobile Deep Neural Networks in the wild](https://mlsanity.com/paper/2109.13963/) (2021)
- [BakedAvatar: Baking Neural Fields for Real-Time Head Avatar Synthesis](https://mlsanity.com/paper/2311.05521/) (2023)
- [Pixel3DMM: Versatile Screen-Space Priors for Single-Image 3D Face Reconstruction](https://mlsanity.com/paper/2505.00615/) (2025)
- [Fast Facial Landmark Detection and Applications: A Survey](https://mlsanity.com/paper/2101.10808/) (2021)
- [GenConViT: Deepfake Video Detection Using Generative Convolutional Vision Transformer](https://mlsanity.com/paper/2307.07036/) (2023)
- [EgoBlur: Responsible Innovation in Aria](https://mlsanity.com/paper/2308.13093/) (2023)
- [A multimodal dataset for understanding the impact of mobile phones on remote online virtual education](https://mlsanity.com/paper/2412.14195/) (2024)
- [Face-LLaVA: Facial Expression and Attribute Understanding through Instruction Tuning](https://mlsanity.com/paper/2504.07198/) (2025)
- [Going Deeper Into Face Detection: A Survey](https://mlsanity.com/paper/2103.14983/) (2021)
- [NanoFlowNet: Real-time Dense Optical Flow on a Nano Quadcopter](https://mlsanity.com/paper/2209.06918/) (2022)
- [AI-Generated Faces in the Real World: A Large-Scale Case Study of Twitter Profile Images](https://mlsanity.com/paper/2404.14244/) (2024)
- [Biometrics and Behavior Analysis for Detecting Distractions in e-Learning](https://mlsanity.com/paper/2405.15434/) (2024)
- [Computational Imaging and Artificial Intelligence: The Next Revolution of Mobile Vision](https://mlsanity.com/paper/2109.08880/) (2021)
- [EResFD: Rediscovery of the Effectiveness of Standard Convolution for Lightweight Face Detection](https://mlsanity.com/paper/2204.01209/) (2022)
- [Unimodal Multi-Task Fusion for Emotional Mimicry Intensity Prediction](https://mlsanity.com/paper/2403.11879/) (2024)
- [Neonatal Face and Facial Landmark Detection from Video Recordings](https://mlsanity.com/paper/2302.04341/) (2023)
- [Block shuffling learning for Deepfake Detection](https://mlsanity.com/paper/2202.02819/) (2022)
- [Gaze-based Attention Recognition for Human-Robot Collaboration](https://mlsanity.com/paper/2303.17619/) (2023)
- [Clothes-Changing Person Re-identification Based On Skeleton Dynamics](https://mlsanity.com/paper/2503.10759/) (2025)
- [EgoEvGesture: Gesture Recognition Based on Egocentric Event Camera](https://mlsanity.com/paper/2503.12419/) (2025)
- [What's wrong with this video? Comparing Explainers for Deepfake Detection](https://mlsanity.com/paper/2105.05902/) (2021)
- [Are Face Detection Models Biased?](https://mlsanity.com/paper/2211.03588/) (2022)
- [Edge Devices Inference Performance Comparison](https://mlsanity.com/paper/2306.12093/) (2023)
- [Disjoint Pose and Shape for 3D Face Reconstruction](https://mlsanity.com/paper/2308.13903/) (2023)
- [Real-time Virtual-Try-On from a Single Example Image through Deep Inverse Graphics and Learned Differentiable Renderers](https://mlsanity.com/paper/2205.06305/) (2022)
- [Real-time Driver Monitoring Systems on Edge AI Device](https://mlsanity.com/paper/2304.01555/) (2023)
- [Evaluation of Video-Based rPPG in Challenging Environments: Artifact Mitigation and Network Resilience](https://mlsanity.com/paper/2405.01230/) (2024)

## Abstract (arXiv)

We present BlazeFace, a lightweight and well-performing face detector tailored for mobile GPU inference. It runs at a speed of 200-1000+ FPS on flagship devices. This super-realtime performance enables it to be applied to any augmented reality pipeline that requires an accurate facial region of interest as an input for task-specific models, such as 2D/3D facial keypoint or geometry estimation, facial features or expression classification, and face region segmentation. Our contributions include a lightweight feature extraction network inspired by, but distinct from MobileNetV1/V2, a GPU-friendly anchor scheme modified from Single Shot MultiBox Detector (SSD), and an improved tie resolution strategy alternative to non-maximum suppression.