Tiny-Face: Ultra-lightweight Face Detection for Mobile Devices
Tiny-Face is an ultra-lightweight face detection model specifically designed to deliver fast and efficient performance on mobile and edge devices, where computational resources are limited. 🚀 Unlike many conventional face detection models, Tiny-Face is streamlined to use minimal memory and processing power while still achieving high precision in detecting faces.
Tiny-Face builds upon the core concepts of RetinaFace but introduces several optimizations that make it ideal for real-world deployment on mobile phones, embedded systems, and IoT devices. With an emphasis on reducing the model’s footprint, Tiny-Face can run seamlessly on low-power hardware without compromising on detection accuracy, making it an excellent choice for applications where speed and efficiency are critical.
👉 https://github.com/yakhyo/tiny-face-pytorch
Key Features 🌟
- Tiny-sized Efficiency: Ultra-lightweight and ideal for mobile or edge devices.
- Mobile-friendly: Slim, RFB, and MobileNet configurations optimized for minimal resources.
- Pretrained Backbones: High precision with models tailored for embedded systems.
📈 Performance on WiderFace Dataset
Multi-scale Image Size
| Models | Pretrained on ImageNet | Easy | Medium | Hard | #Params(M) | Size(MB) |
|---|---|---|---|---|---|---|
| SlimFace | False | 79.50% | 79.40% | 68.36% | 0.343 | 1.4 |
| RFB | False | 80.49% | 81.51% | 75.73% | 0.359 | 1.5 |
| RetinaFace | True | 87.69% | 86.39% | 80.21% | 0.426 | 1.8 |
Original Image Size
| Models | Pretrained on ImageNet | Easy | Medium | Hard | #Params(M) |
|---|---|---|---|---|---|
| SlimFace | False | 87.10% | 84.36% | 67.38% | 0.343 |
| RFB | False | 87.09% | 84.61% | 69.22% | 0.359 |
| RetinaFace | True | 90.26% | 87.48% | 72.85% | 0.426 |
Explore the https://github.com/yakhyo/tiny-face-pytorch for detailed setup instructions and contribute to this efficient face detection project!