Hugging Face's logo

qualcomm
/
Facial-Landmark-Detection

Keypoint Detection
PyTorch
LiteRT
ONNX
backbone
android
Model card Files Community

Facial-Landmark-Detection: Optimized for Mobile Deployment

Real-time 3D facial landmark detection optimized for mobile and edge

Detects facial landmarks (eg, nose, mouth, etc.). This model's architecture was developed by Qualcomm. The model was trained by Qualcomm on a proprietary dataset of faces, but can be used on any image.

This repository provides scripts to run Facial-Landmark-Detection on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Model_use_case.pose_estimation
  • Model Stats:
    • Input resolution: 128x128
    • Number of parameters: 5.424M
    • Model size (float): 21.256MB
    • Model size (w8a8): 5.314MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
Facial-Landmark-Detection float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 6.286 ms 0 - 13 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 6.271 ms 0 - 10 MB NPU Use Export Script
Facial-Landmark-Detection float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 0.386 ms 0 - 31 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 0.469 ms 0 - 19 MB NPU Use Export Script
Facial-Landmark-Detection float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 0.282 ms 0 - 100 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 0.274 ms 0 - 3 MB NPU Use Export Script
Facial-Landmark-Detection float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 0.502 ms 0 - 15 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 0.475 ms 0 - 12 MB NPU Use Export Script
Facial-Landmark-Detection float SA7255P ADP Qualcomm® SA7255P TFLITE 6.286 ms 0 - 13 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float SA7255P ADP Qualcomm® SA7255P QNN 6.271 ms 0 - 10 MB NPU Use Export Script
Facial-Landmark-Detection float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 0.279 ms 0 - 100 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 0.273 ms 0 - 11 MB NPU Use Export Script
Facial-Landmark-Detection float SA8295P ADP Qualcomm® SA8295P TFLITE 0.637 ms 0 - 18 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float SA8295P ADP Qualcomm® SA8295P QNN 0.624 ms 0 - 17 MB NPU Use Export Script
Facial-Landmark-Detection float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 0.28 ms 0 - 101 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 0.279 ms 0 - 2 MB NPU Use Export Script
Facial-Landmark-Detection float SA8775P ADP Qualcomm® SA8775P TFLITE 0.502 ms 0 - 15 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float SA8775P ADP Qualcomm® SA8775P QNN 0.475 ms 0 - 12 MB NPU Use Export Script
Facial-Landmark-Detection float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 0.285 ms 0 - 101 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 0.282 ms 0 - 49 MB NPU Use Export Script
Facial-Landmark-Detection float Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 0.388 ms 0 - 30 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 0.209 ms 0 - 32 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 0.226 ms 8 - 28 MB NPU Use Export Script
Facial-Landmark-Detection float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 0.3 ms 0 - 26 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 0.216 ms 0 - 14 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 0.191 ms 0 - 14 MB NPU Use Export Script
Facial-Landmark-Detection float Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 0.272 ms 0 - 15 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection float Snapdragon X Elite CRD Snapdragon® X Elite QNN 0.33 ms 0 - 0 MB NPU Use Export Script
Facial-Landmark-Detection float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 0.359 ms 10 - 10 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 0.451 ms 0 - 11 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN 1.07 ms 0 - 9 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 0.215 ms 0 - 27 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN 0.233 ms 0 - 34 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 0.171 ms 0 - 43 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN 0.164 ms 0 - 3 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 0.336 ms 0 - 13 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN 0.315 ms 0 - 12 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) TFLITE 0.553 ms 0 - 22 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) QNN 0.612 ms 0 - 14 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 1.94 ms 0 - 11 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 0.451 ms 0 - 11 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 SA7255P ADP Qualcomm® SA7255P QNN 1.07 ms 0 - 9 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 0.167 ms 0 - 43 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN 0.161 ms 0 - 2 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 0.44 ms 0 - 15 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 SA8295P ADP Qualcomm® SA8295P QNN 0.446 ms 0 - 17 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 0.168 ms 0 - 43 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN 0.155 ms 0 - 2 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 0.336 ms 0 - 13 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 SA8775P ADP Qualcomm® SA8775P QNN 0.315 ms 0 - 12 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile TFLITE 0.174 ms 0 - 42 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile QNN 0.165 ms 0 - 43 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 Samsung Galaxy S23 Snapdragon® 8 Gen 2 Mobile ONNX 0.448 ms 0 - 18 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 0.14 ms 0 - 32 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN 0.131 ms 0 - 25 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 0.33 ms 0 - 40 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection w8a8 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile TFLITE 0.129 ms 0 - 17 MB NPU Facial-Landmark-Detection.tflite
Facial-Landmark-Detection w8a8 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile QNN 0.126 ms 0 - 17 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 Snapdragon 8 Elite QRD Snapdragon® 8 Elite Mobile ONNX 0.381 ms 0 - 20 MB NPU Facial-Landmark-Detection.onnx
Facial-Landmark-Detection w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN 0.23 ms 0 - 0 MB NPU Use Export Script
Facial-Landmark-Detection w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 0.475 ms 4 - 4 MB NPU Facial-Landmark-Detection.onnx

Installation

Install the package via pip:

pip install "qai-hub-models[facemap-3dmm]"

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.facemap_3dmm.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.facemap_3dmm.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.facemap_3dmm.export

Profiling Results
------------------------------------------------------------
Facial-Landmark-Detection
Device                          : cs_8275 (ANDROID 14)                
Runtime                         : TFLITE                              
Estimated inference time (ms)   : 6.3                                 
Estimated peak memory usage (MB): [0, 13]                             
Total # Ops                     : 37                                  
Compute Unit(s)                 : npu (37 ops) gpu (0 ops) cpu (0 ops)

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.facemap_3dmm import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S24")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.facemap_3dmm.demo --eval-mode on-device

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.facemap_3dmm.demo -- --eval-mode on-device

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 Facial-Landmark-Detection's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of Facial-Landmark-Detection can be found here.
  • The license for the compiled assets for on-device deployment can be found here

Community

Downloads last month
79
Inference Providers NEW
Keypoint Detection
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support