MediaPipe-Face-Detection-Quantized: Optimized for Mobile Deployment
Detect faces and locate facial features in real-time video and image streams
Designed for sub-millisecond processing, this model predicts bounding boxes and pose skeletons (left eye, right eye, nose tip, mouth, left eye tragion, and right eye tragion) of faces in an image.
This model is an implementation of MediaPipe-Face-Detection-Quantized found here.
This repository provides scripts to run MediaPipe-Face-Detection-Quantized on Qualcomm® devices. More details on model performance across various devices, can be found here.
Model Details
- Model Type: Object detection
- Model Stats:
- Input resolution: 256x256
- Number of output classes: 6
- Number of parameters (MediaPipeFaceDetector): 135K
- Model size (MediaPipeFaceDetector): 255 KB
- Number of parameters (MediaPipeFaceLandmarkDetector): 603K
- Model size (MediaPipeFaceLandmarkDetector): 746 KB
Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model |
---|---|---|---|---|---|---|---|---|
MediaPipeFaceDetector | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 0.274 ms | 0 - 73 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.304 ms | 0 - 73 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.so |
MediaPipeFaceDetector | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.19 ms | 0 - 17 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.206 ms | 0 - 17 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.so |
MediaPipeFaceDetector | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.167 ms | 0 - 14 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.175 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 0.765 ms | 0 - 19 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 0.827 ms | 0 - 8 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 5.221 ms | 0 - 5 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 0.275 ms | 0 - 10 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.306 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | SA7255P ADP | SA7255P | TFLITE | 2.123 ms | 0 - 16 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | SA7255P ADP | SA7255P | QNN | 2.267 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 0.273 ms | 0 - 5 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.307 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | SA8295P ADP | SA8295P | TFLITE | 0.664 ms | 0 - 14 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | SA8295P ADP | SA8295P | QNN | 0.749 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 0.272 ms | 0 - 5 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.305 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | SA8775P ADP | SA8775P | TFLITE | 0.617 ms | 0 - 14 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | SA8775P ADP | SA8775P | QNN | 0.813 ms | 0 - 5 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 0.321 ms | 0 - 19 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceDetector | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.363 ms | 0 - 20 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceDetector | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.419 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 0.186 ms | 0 - 4 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.22 ms | 0 - 10 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.so |
MediaPipeFaceLandmarkDetector | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.129 ms | 0 - 13 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.162 ms | 0 - 12 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.so |
MediaPipeFaceLandmarkDetector | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.141 ms | 0 - 10 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.171 ms | 0 - 10 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 0.406 ms | 0 - 12 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 0.498 ms | 0 - 8 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 2.963 ms | 0 - 6 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 0.18 ms | 0 - 3 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.221 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | SA7255P ADP | SA7255P | TFLITE | 0.997 ms | 0 - 10 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | SA7255P ADP | SA7255P | QNN | 1.197 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 0.18 ms | 0 - 9 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.222 ms | 0 - 1 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | SA8295P ADP | SA8295P | TFLITE | 0.482 ms | 0 - 9 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | SA8295P ADP | SA8295P | QNN | 0.69 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 0.187 ms | 0 - 10 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.221 ms | 0 - 2 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | SA8775P ADP | SA8775P | TFLITE | 0.445 ms | 0 - 8 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | SA8775P ADP | SA8775P | QNN | 0.63 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 0.224 ms | 0 - 14 MB | FP16 | NPU | MediaPipe-Face-Detection-Quantized.tflite |
MediaPipeFaceLandmarkDetector | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.261 ms | 0 - 15 MB | FP16 | NPU | Use Export Script |
MediaPipeFaceLandmarkDetector | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.337 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
Installation
This model can be installed as a Python package via pip.
pip install "qai-hub-models[mediapipe_face_quantized]"
Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
Sign-in to Qualcomm® AI Hub with your
Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token
.
With this API token, you can configure your client to run models on the cloud hosted devices.
qai-hub configure --api_token API_TOKEN
Navigate to docs for more information.
Demo off target
The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.
python -m qai_hub_models.models.mediapipe_face_quantized.demo
The above demo runs a reference implementation of pre-processing, model inference, and post processing.
NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).
%run -m qai_hub_models.models.mediapipe_face_quantized.demo
Run model on a cloud-hosted device
In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:
- Performance check on-device on a cloud-hosted device
- Downloads compiled assets that can be deployed on-device for Android.
- Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.mediapipe_face_quantized.export
Profiling Results
------------------------------------------------------------
MediaPipeFaceDetector
Device : Samsung Galaxy S23 (13)
Runtime : TFLITE
Estimated inference time (ms) : 0.3
Estimated peak memory usage (MB): [0, 73]
Total # Ops : 121
Compute Unit(s) : NPU (121 ops)
------------------------------------------------------------
MediaPipeFaceLandmarkDetector
Device : Samsung Galaxy S23 (13)
Runtime : TFLITE
Estimated inference time (ms) : 0.2
Estimated peak memory usage (MB): [0, 4]
Total # Ops : 117
Compute Unit(s) : NPU (117 ops)
Deploying compiled model to Android
The models can be deployed using multiple runtimes:
TensorFlow Lite (
.tflite
export): This tutorial provides a guide to deploy the .tflite model in an Android application.QNN (
.so
export ): This sample app provides instructions on how to use the.so
shared library in an Android application.
View on Qualcomm® AI Hub
Get more details on MediaPipe-Face-Detection-Quantized's performance across various devices here. Explore all available models on Qualcomm® AI Hub
License
- The license for the original implementation of MediaPipe-Face-Detection-Quantized can be found here.
- The license for the compiled assets for on-device deployment can be found here
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.