Hugging Face's logo

qualcomm
/
Yolo-v7

Object Detection
PyTorch
real_time
android
Model card Files
xet
Community

Yolo-v7: Optimized for Mobile Deployment

Real-time object detection optimized for mobile and edge

YoloV7 is a machine learning model that predicts bounding boxes and classes of objects in an image.

This model is an implementation of Yolo-v7 found here.

This repository provides scripts to run Yolo-v7 on Qualcomm® devices. More details on model performance across various devices, can be found here.

WARNING: The model assets are not readily available for download due to licensing restrictions.

Model Details

  • Model Type: Model_use_case.object_detection
  • Model Stats:
    • Model checkpoint: YoloV7 Tiny
    • Input resolution: 640x640
    • Number of parameters: 6.24M
    • Model size (float): 23.8 MB
    • Model size (w8a8): 6.23 MB
    • Model size (w8a16): 6.66 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
Yolo-v7 float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 23.451 ms 1 - 31 MB NPU --
Yolo-v7 float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 18.901 ms 0 - 155 MB NPU --
Yolo-v7 float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 11.124 ms 1 - 47 MB NPU --
Yolo-v7 float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 10.619 ms 4 - 42 MB NPU --
Yolo-v7 float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 8.963 ms 1 - 9 MB NPU --
Yolo-v7 float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 6.462 ms 0 - 162 MB NPU --
Yolo-v7 float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 8.207 ms 0 - 107 MB NPU --
Yolo-v7 float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 10.759 ms 0 - 31 MB NPU --
Yolo-v7 float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 33.361 ms 1 - 145 MB NPU --
Yolo-v7 float SA7255P ADP Qualcomm® SA7255P TFLITE 23.451 ms 1 - 31 MB NPU --
Yolo-v7 float SA7255P ADP Qualcomm® SA7255P QNN_DLC 18.901 ms 0 - 155 MB NPU --
Yolo-v7 float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 8.92 ms 0 - 29 MB NPU --
Yolo-v7 float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 6.451 ms 0 - 163 MB NPU --
Yolo-v7 float SA8295P ADP Qualcomm® SA8295P TFLITE 12.882 ms 1 - 39 MB NPU --
Yolo-v7 float SA8295P ADP Qualcomm® SA8295P QNN_DLC 9.163 ms 0 - 43 MB NPU --
Yolo-v7 float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 8.901 ms 1 - 13 MB NPU --
Yolo-v7 float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 6.456 ms 0 - 168 MB NPU --
Yolo-v7 float SA8775P ADP Qualcomm® SA8775P TFLITE 10.759 ms 0 - 31 MB NPU --
Yolo-v7 float SA8775P ADP Qualcomm® SA8775P QNN_DLC 33.361 ms 1 - 145 MB NPU --
Yolo-v7 float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 6.42 ms 0 - 44 MB NPU --
Yolo-v7 float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 4.671 ms 5 - 357 MB NPU --
Yolo-v7 float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 5.589 ms 5 - 347 MB NPU --
Yolo-v7 float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 5.529 ms 1 - 36 MB NPU --
Yolo-v7 float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 3.595 ms 5 - 133 MB NPU --
Yolo-v7 float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 5.435 ms 0 - 135 MB NPU --
Yolo-v7 float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 4.952 ms 0 - 35 MB NPU --
Yolo-v7 float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 3.015 ms 5 - 156 MB NPU --
Yolo-v7 float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 3.787 ms 1 - 138 MB NPU --
Yolo-v7 float Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 6.827 ms 216 - 216 MB NPU --
Yolo-v7 float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 8.445 ms 8 - 8 MB NPU --
Yolo-v7 w8a16 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 QNN_DLC 11.052 ms 2 - 112 MB NPU --
Yolo-v7 w8a16 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 ONNX 427.889 ms 83 - 88 MB CPU --
Yolo-v7 w8a16 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 8.972 ms 2 - 192 MB NPU --
Yolo-v7 w8a16 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 6.257 ms 2 - 204 MB NPU --
Yolo-v7 w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 3.927 ms 2 - 15 MB NPU --
Yolo-v7 w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 5.25 ms 0 - 26 MB NPU --
Yolo-v7 w8a16 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 19.033 ms 2 - 195 MB NPU --
Yolo-v7 w8a16 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 180.561 ms 81 - 89 MB CPU --
Yolo-v7 w8a16 SA7255P ADP Qualcomm® SA7255P QNN_DLC 8.972 ms 2 - 192 MB NPU --
Yolo-v7 w8a16 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 3.924 ms 2 - 17 MB NPU --
Yolo-v7 w8a16 SA8295P ADP Qualcomm® SA8295P QNN_DLC 5.653 ms 2 - 193 MB NPU --
Yolo-v7 w8a16 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 3.929 ms 2 - 17 MB NPU --
Yolo-v7 w8a16 SA8775P ADP Qualcomm® SA8775P QNN_DLC 19.033 ms 2 - 195 MB NPU --
Yolo-v7 w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 2.832 ms 2 - 201 MB NPU --
Yolo-v7 w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 3.505 ms 1 - 219 MB NPU --
Yolo-v7 w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 2.065 ms 2 - 197 MB NPU --
Yolo-v7 w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 2.678 ms 0 - 196 MB NPU --
Yolo-v7 w8a16 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile QNN_DLC 4.984 ms 2 - 129 MB NPU --
Yolo-v7 w8a16 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile ONNX 202.192 ms 82 - 99 MB CPU --
Yolo-v7 w8a16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 1.62 ms 2 - 196 MB NPU --
Yolo-v7 w8a16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 2.153 ms 1 - 196 MB NPU --
Yolo-v7 w8a16 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 4.378 ms 81 - 81 MB NPU --
Yolo-v7 w8a16 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 5.39 ms 5 - 5 MB NPU --
Yolo-v7 w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 TFLITE 5.903 ms 0 - 9 MB NPU --
Yolo-v7 w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 QNN_DLC 6.106 ms 1 - 106 MB NPU --
Yolo-v7 w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 ONNX 72.271 ms 38 - 45 MB CPU --
Yolo-v7 w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 4.875 ms 0 - 28 MB NPU --
Yolo-v7 w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 4.6 ms 1 - 44 MB NPU --
Yolo-v7 w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 2.885 ms 0 - 45 MB NPU --
Yolo-v7 w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 2.852 ms 1 - 57 MB NPU --
Yolo-v7 w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 2.189 ms 0 - 34 MB NPU --
Yolo-v7 w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 2.129 ms 1 - 15 MB NPU --
Yolo-v7 w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 2.736 ms 0 - 24 MB NPU --
Yolo-v7 w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 2.67 ms 0 - 27 MB NPU --
Yolo-v7 w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 2.529 ms 1 - 45 MB NPU --
Yolo-v7 w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 77.063 ms 3 - 46 MB GPU --
Yolo-v7 w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 53.787 ms 36 - 46 MB CPU --
Yolo-v7 w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 4.875 ms 0 - 28 MB NPU --
Yolo-v7 w8a8 SA7255P ADP Qualcomm® SA7255P QNN_DLC 4.6 ms 1 - 44 MB NPU --
Yolo-v7 w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 2.208 ms 0 - 33 MB NPU --
Yolo-v7 w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 2.136 ms 1 - 10 MB NPU --
Yolo-v7 w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 3.655 ms 0 - 35 MB NPU --
Yolo-v7 w8a8 SA8295P ADP Qualcomm® SA8295P QNN_DLC 3.483 ms 1 - 52 MB NPU --
Yolo-v7 w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 2.209 ms 0 - 34 MB NPU --
Yolo-v7 w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 2.135 ms 1 - 12 MB NPU --
Yolo-v7 w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 2.67 ms 0 - 27 MB NPU --
Yolo-v7 w8a8 SA8775P ADP Qualcomm® SA8775P QNN_DLC 2.529 ms 1 - 45 MB NPU --
Yolo-v7 w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 1.499 ms 0 - 46 MB NPU --
Yolo-v7 w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 1.413 ms 1 - 62 MB NPU --
Yolo-v7 w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 1.741 ms 0 - 63 MB NPU --
Yolo-v7 w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 1.221 ms 0 - 38 MB NPU --
Yolo-v7 w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 1.048 ms 1 - 50 MB NPU --
Yolo-v7 w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 1.383 ms 0 - 53 MB NPU --
Yolo-v7 w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile TFLITE 2.462 ms 0 - 39 MB NPU --
Yolo-v7 w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile QNN_DLC 2.726 ms 1 - 194 MB NPU --
Yolo-v7 w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile ONNX 50.774 ms 41 - 58 MB CPU --
Yolo-v7 w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 1.049 ms 0 - 35 MB NPU --
Yolo-v7 w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 0.888 ms 1 - 52 MB NPU --
Yolo-v7 w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 1.254 ms 0 - 54 MB NPU --
Yolo-v7 w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 2.418 ms 20 - 20 MB NPU --
Yolo-v7 w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 2.748 ms 5 - 5 MB NPU --

Installation

Install the package via pip:

# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[yolov7]"

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.yolov7.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.yolov7.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.yolov7.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.yolov7 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.yolov7.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.yolov7.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 Yolo-v7's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of Yolo-v7 can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Object Detection
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support