Hugging Face's logo

qualcomm
/
MobileFaceNet

Object Detection
PyTorch
real_time
bu_auto
android
Model card Files
xet
 Community
MobileFaceNet / README.md
qaihm-bot's picture
qaihm-bot
v0.64.0
c8eeb67 verified
preview code
|
Raw History Blame Contribute Delete
11.6 kB
metadata 
library_name: pytorch
license: apache-2.0
tags:
  - real_time
  - bu_auto
  - android
pipeline_tag: object-detection

MobileFaceNet: Optimized for Qualcomm Devices

MobileFaceNet is an efficient CNN that maps a 112x112 face image to a compact 128-dimensional embedding. Two embeddings are compared via cosine similarity to determine whether they belong to the same person, achieving 99.48% accuracy on the LFW benchmark. The model uses depthwise-separable convolutions and inverted residual blocks (MobileNetV2-style) to stay under 1M parameters, making it well-suited for real-time face verification on mobile and edge devices. Trained with ArcFace loss on MS-Celeb-1M.

This is based on the implementation of MobileFaceNet found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime Precision Chipset SDK Versions Download
ONNX float Universal QAIRT 2.50, ONNX Runtime 1.30.0 Download
ONNX w8a16 Universal QAIRT 2.50, ONNX Runtime 1.30.0 Download
QNN_DLC float Universal QAIRT 2.50 Download
QNN_DLC w8a16 Universal QAIRT 2.50 Download
TFLITE float Universal QAIRT 2.50 Download

For more device-specific assets and performance metrics, visit MobileFaceNet on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

  • Custom weights (e.g., fine-tuned checkpoints)
  • Custom input shapes
  • Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for MobileFaceNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

  • Embedding dimension: 128
  • Input resolution: 112x112
  • Model checkpoint: mobilefacenet.pt
  • Model size (float): 4MB
  • Number of parameters: 1M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
MobileFaceNet ONNX float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.538 ms 0 - 36 MB NPU
MobileFaceNet ONNX float Snapdragon® 8 Elite For Galaxy Mobile 0.546 ms 0 - 36 MB NPU
MobileFaceNet ONNX float Snapdragon® X2 Elite 0.611 ms 1 - 1 MB NPU
MobileFaceNet ONNX float Snapdragon® X Elite 1.038 ms 0 - 0 MB NPU
MobileFaceNet ONNX float Snapdragon® 8 Gen 3 Mobile 0.704 ms 0 - 57 MB NPU
MobileFaceNet ONNX float Snapdragon® 8 Gen 1 Mobile 1.478 ms 0 - 60 MB NPU
MobileFaceNet ONNX float Qualcomm® Dragonwing™ IQ-8275 1.449 ms 0 - 4 MB NPU
MobileFaceNet ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.0 ms 0 - 5 MB NPU
MobileFaceNet ONNX float Qualcomm® QCS8450 1.478 ms 0 - 60 MB NPU
MobileFaceNet ONNX float Qualcomm® Dragonwing™ IQ-9075 1.382 ms 0 - 4 MB NPU
MobileFaceNet ONNX float Qualcomm® Dragonwing™ IQ-X7181 1.038 ms 0 - 0 MB NPU
MobileFaceNet ONNX float Qualcomm® Dragonwing™ Q-8750 0.546 ms 0 - 36 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.373 ms 0 - 47 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.451 ms 0 - 50 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® X2 Elite 0.457 ms 1 - 1 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® X Elite 0.817 ms 0 - 0 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.564 ms 0 - 60 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 1.09 ms 0 - 62 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 2.325 ms 0 - 3 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 0.852 ms 0 - 4 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.794 ms 0 - 5 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® QCS8450 1.09 ms 0 - 62 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 0.917 ms 0 - 3 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 0.817 ms 0 - 0 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 5.412 ms 0 - 158 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 0.943 ms 0 - 47 MB NPU
MobileFaceNet ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.451 ms 0 - 50 MB NPU
MobileFaceNet ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 0.943 ms 0 - 47 MB NPU
MobileFaceNet QNN_DLC float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.519 ms 0 - 35 MB NPU
MobileFaceNet QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 0.591 ms 0 - 40 MB NPU
MobileFaceNet QNN_DLC float Snapdragon® X2 Elite 0.719 ms 0 - 0 MB NPU
MobileFaceNet QNN_DLC float Snapdragon® X Elite 1.324 ms 0 - 0 MB NPU
MobileFaceNet QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.779 ms 0 - 52 MB NPU
MobileFaceNet QNN_DLC float Snapdragon® 8 Gen 1 Mobile 1.623 ms 0 - 55 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 1.472 ms 0 - 3 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.169 ms 1 - 3 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® SA8775P 1.613 ms 0 - 39 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® SA8650P 1.613 ms 0 - 39 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® SA8255P 1.613 ms 0 - 39 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® QCS8450 1.623 ms 0 - 55 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.475 ms 0 - 3 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.324 ms 0 - 0 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.591 ms 0 - 40 MB NPU
MobileFaceNet QNN_DLC float Qualcomm® SA8295P 1.916 ms 0 - 33 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.377 ms 0 - 42 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.493 ms 0 - 45 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® X2 Elite 0.59 ms 0 - 0 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® X Elite 1.141 ms 0 - 0 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 0.679 ms 0 - 53 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 1.237 ms 0 - 59 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS6490 2.751 ms 2 - 4 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 0.998 ms 0 - 3 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.994 ms 0 - 2 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® SA8775P 1.201 ms 0 - 43 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® SA8650P 1.201 ms 0 - 43 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® SA8255P 1.201 ms 0 - 43 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® QCS8450 1.237 ms 0 - 59 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 1.088 ms 2 - 4 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.141 ms 0 - 0 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 6.102 ms 19 - 167 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 1.079 ms 0 - 41 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.493 ms 0 - 45 MB NPU
MobileFaceNet QNN_DLC w8a16 Qualcomm® SA8295P 1.499 ms 0 - 40 MB NPU
MobileFaceNet QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 1.079 ms 0 - 41 MB NPU
MobileFaceNet TFLITE float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.537 ms 0 - 36 MB NPU
MobileFaceNet TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 0.56 ms 0 - 35 MB NPU
MobileFaceNet TFLITE float Snapdragon® 8 Gen 3 Mobile 0.697 ms 0 - 52 MB NPU
MobileFaceNet TFLITE float Snapdragon® 8 Gen 1 Mobile 1.412 ms 0 - 55 MB NPU
MobileFaceNet TFLITE float Qualcomm® Dragonwing™ IQ-8275 1.44 ms 0 - 6 MB NPU
MobileFaceNet TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.98 ms 0 - 2 MB NPU
MobileFaceNet TFLITE float Qualcomm® SA8775P 1.488 ms 0 - 40 MB NPU
MobileFaceNet TFLITE float Qualcomm® SA8650P 1.488 ms 0 - 40 MB NPU
MobileFaceNet TFLITE float Qualcomm® SA8255P 1.488 ms 0 - 40 MB NPU
MobileFaceNet TFLITE float Qualcomm® QCS8450 1.412 ms 0 - 55 MB NPU
MobileFaceNet TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.342 ms 0 - 5 MB NPU
MobileFaceNet TFLITE float Qualcomm® Dragonwing™ Q-8750 0.56 ms 0 - 35 MB NPU
MobileFaceNet TFLITE float Qualcomm® SA8295P 1.705 ms 0 - 33 MB NPU

License

  • The license for the original implementation of MobileFaceNet can be found here.

References

Community