Person-Foot-Detection: Optimized for Mobile Deployment

Multi-task Human detector

Real-time multiple person detection with accurate feet localization optimized for mobile and edge.

This model is an implementation of Person-Foot-Detection found here.

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

Model Details

Model Type: Model_use_case.object_detection
Model Stats:
- Inference latency: RealTime
- Input resolution: 640x480
- Number of output classes: 2
- Number of parameters: 2.53M
- Model size (float): 9.69 MB
- Model size (w8a8): 2.62 MB
- Model size (w8a16): 2.90 MB

Model	Precision	Device	Chipset	Target Runtime	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit	Target Model
Person-Foot-Detection	float	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	TFLITE	23.358 ms	5 - 32 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	22.535 ms	3 - 30 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	TFLITE	7.065 ms	5 - 43 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	12.261 ms	4 - 42 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	TFLITE	5.06 ms	5 - 18 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	4.479 ms	4 - 15 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	6.134 ms	10 - 27 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	TFLITE	7.943 ms	5 - 33 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	7.131 ms	2 - 28 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	SA7255P ADP	Qualcomm® SA7255P	TFLITE	23.358 ms	5 - 32 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	22.535 ms	3 - 30 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	TFLITE	5.14 ms	1 - 9 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	QNN_DLC	4.468 ms	2 - 13 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	SA8295P ADP	Qualcomm® SA8295P	TFLITE	8.407 ms	5 - 39 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	8.805 ms	0 - 37 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	TFLITE	5.2 ms	5 - 15 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	QNN_DLC	4.483 ms	4 - 16 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	SA8775P ADP	Qualcomm® SA8775P	TFLITE	7.943 ms	5 - 33 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	7.131 ms	2 - 28 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	TFLITE	3.415 ms	0 - 39 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	3.164 ms	4 - 37 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	4.73 ms	1 - 40 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	TFLITE	2.93 ms	0 - 32 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	2.55 ms	4 - 41 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	3.828 ms	1 - 34 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	TFLITE	2.796 ms	0 - 33 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_DLC	2.409 ms	4 - 77 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	ONNX	2.632 ms	3 - 39 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	4.89 ms	4 - 4 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	float	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	5.608 ms	17 - 17 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	QNN_DLC	12.709 ms	2 - 123 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	ONNX	538.772 ms	93 - 97 MB	CPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	8.348 ms	2 - 32 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	5.555 ms	2 - 41 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	3.312 ms	2 - 10 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	4.754 ms	5 - 21 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	19.62 ms	2 - 33 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	RB5 (Proxy)	Qualcomm® QCS8250 (Proxy)	ONNX	232.379 ms	82 - 86 MB	CPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	8.348 ms	2 - 32 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	QNN_DLC	3.313 ms	2 - 10 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	5.031 ms	2 - 40 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	QNN_DLC	3.33 ms	2 - 10 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	19.62 ms	2 - 33 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	2.32 ms	2 - 39 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	3.234 ms	0 - 41 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	1.877 ms	2 - 38 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	2.897 ms	0 - 41 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	QNN_DLC	7.809 ms	2 - 38 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	ONNX	276.973 ms	94 - 109 MB	CPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_DLC	2.054 ms	2 - 37 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	ONNX	2.754 ms	1 - 41 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a16	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	3.747 ms	0 - 0 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a16	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	4.769 ms	10 - 10 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	TFLITE	4.619 ms	1 - 8 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	QNN_DLC	4.729 ms	1 - 102 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	ONNX	76.275 ms	50 - 58 MB	CPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	TFLITE	3.659 ms	0 - 27 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	3.55 ms	1 - 29 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	TFLITE	1.402 ms	0 - 35 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	1.437 ms	1 - 39 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	TFLITE	1.149 ms	0 - 6 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	1.172 ms	1 - 12 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	1.675 ms	0 - 14 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	TFLITE	1.64 ms	0 - 27 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	1.556 ms	1 - 29 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	RB5 (Proxy)	Qualcomm® QCS8250 (Proxy)	TFLITE	27.487 ms	1 - 4 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	RB5 (Proxy)	Qualcomm® QCS8250 (Proxy)	ONNX	58.61 ms	48 - 54 MB	CPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	SA7255P ADP	Qualcomm® SA7255P	TFLITE	3.659 ms	0 - 27 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	3.55 ms	1 - 29 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	TFLITE	1.153 ms	0 - 13 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	QNN_DLC	1.166 ms	1 - 13 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	SA8295P ADP	Qualcomm® SA8295P	TFLITE	2.236 ms	0 - 33 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	2.236 ms	0 - 34 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	TFLITE	1.189 ms	0 - 13 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	QNN_DLC	1.171 ms	1 - 12 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	SA8775P ADP	Qualcomm® SA8775P	TFLITE	1.64 ms	0 - 27 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	1.556 ms	1 - 29 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	TFLITE	0.793 ms	0 - 44 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	0.798 ms	1 - 41 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	1.099 ms	0 - 42 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	TFLITE	0.751 ms	0 - 30 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	0.637 ms	1 - 35 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	0.953 ms	0 - 32 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	TFLITE	1.825 ms	0 - 35 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	QNN_DLC	2.952 ms	1 - 36 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	ONNX	71.533 ms	51 - 68 MB	CPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	TFLITE	0.653 ms	0 - 29 MB	NPU	Person-Foot-Detection.tflite
Person-Foot-Detection	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_DLC	0.517 ms	1 - 39 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	ONNX	0.794 ms	0 - 38 MB	NPU	Person-Foot-Detection.onnx.zip
Person-Foot-Detection	w8a8	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	1.393 ms	0 - 0 MB	NPU	Person-Foot-Detection.dlc
Person-Foot-Detection	w8a8	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	1.617 ms	8 - 8 MB	NPU	Person-Foot-Detection.onnx.zip

Installation

Install the package via pip:

# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[foot-track-net]"

Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub Workbench 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.foot_track_net.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.foot_track_net.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.foot_track_net.export

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.foot_track_net import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# 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 Workbench. 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.foot_track_net.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.foot_track_net.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 Person-Foot-Detection's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

The license for the original implementation of Person-Foot-Detection can be found here.
The license for the compiled assets for on-device deployment can be found here

References

Source Model Implementation

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.

Downloads last month: 183