SlimFace Demonstration

SlimFace: Advanced Face Classification with TorchVision Backbones

SlimFace empowers developers to build high-accuracy face classification models by leveraging TorchVision's powerful pre-trained architectures through a transfer learning approach. 🌟 It provides a flexible, efficient, and scalable solution for facial recognition, supporting training on custom datasets and inference to deliver top-tier performance for custom applications.

For more information, you can follow below:
Source code: , Author: , PyTorch Docs:

Upload Image

Model Files

Reference Dict JSON

Index to Class Mapping JSON

Classifier Model (.pth)

EdgeFace Model (.pt)

Advanced Settings

Detection Algorithm

Accelerator

Image Resolution

128 512

Similarity Threshold

0.1 1

Example Images

example_face_1.jpg

example_face_2.jpg

example_face_3.jpg

example_face_4.jpg

More Information about SlimFace

This project leverages code from by , with our own bug fixes and enhancements available at .

Supported Architectures:

EfficientNet: B0-B7 and V2 (Small, Medium, Large) for balanced performance and efficiency. 📸
RegNet: X/Y series (400MF to 128GF) for optimized computation across diverse hardware. 💻
Vision Transformers (ViT): B_16, B_32, H_14, L_16, L_32 for cutting-edge feature extraction. 🚀