Hugging Face's logo

qualcomm
/
FFNet-78S-LowRes

Image Segmentation
PyTorch
LiteRT
real_time
android
Model card Files
xet
Community

FFNet-78S-LowRes: Optimized for Mobile Deployment

Semantic segmentation for automotive street scenes

FFNet-78S-LowRes is a "fuss-free network" that segments street scene images with per-pixel classes like road, sidewalk, and pedestrian. Trained on the Cityscapes dataset.

This model is an implementation of FFNet-78S-LowRes found here.

This repository provides scripts to run FFNet-78S-LowRes on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Model_use_case.semantic_segmentation
  • Model Stats:
    • Model checkpoint: ffnet78S_BCC_cityscapes_state_dict_quarts_pre_down
    • Input resolution: 1024x512
    • Number of output classes: 19
    • Number of parameters: 26.8M
    • Model size (float): 102 MB
    • Model size (w8a8): 26.0 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
FFNet-78S-LowRes float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 48.925 ms 1 - 61 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 49.519 ms 0 - 28 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 16.357 ms 1 - 116 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 23.941 ms 2 - 40 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 12.608 ms 0 - 352 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 13.464 ms 6 - 19 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 7.981 ms 0 - 120 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 17.153 ms 1 - 62 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 17.855 ms 2 - 30 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float SA7255P ADP Qualcomm® SA7255P TFLITE 48.925 ms 1 - 61 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float SA7255P ADP Qualcomm® SA7255P QNN_DLC 49.519 ms 0 - 28 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 12.586 ms 0 - 353 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 13.421 ms 6 - 19 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float SA8295P ADP Qualcomm® SA8295P TFLITE 18.872 ms 1 - 62 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float SA8295P ADP Qualcomm® SA8295P QNN_DLC 19.764 ms 3 - 31 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 12.619 ms 0 - 322 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 13.483 ms 3 - 23 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float SA8775P ADP Qualcomm® SA8775P TFLITE 17.153 ms 1 - 62 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float SA8775P ADP Qualcomm® SA8775P QNN_DLC 17.855 ms 2 - 30 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 8.557 ms 0 - 117 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 8.971 ms 6 - 42 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 5.727 ms 6 - 53 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 7.296 ms 1 - 66 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 7.54 ms 6 - 40 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 7.265 ms 3 - 42 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 6.175 ms 1 - 66 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 6.698 ms 6 - 41 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 4.017 ms 5 - 46 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes float Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 14.111 ms 113 - 113 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 8.153 ms 47 - 47 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 7.753 ms 0 - 41 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 10.569 ms 2 - 44 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 3.359 ms 0 - 76 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 7.148 ms 2 - 76 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 2.971 ms 0 - 197 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 4.628 ms 1 - 178 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 3.07 ms 0 - 67 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 3.452 ms 0 - 41 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 5.118 ms 2 - 44 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) TFLITE 120.144 ms 7 - 39 MB GPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) QNN_DLC 22.285 ms 2 - 65 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) ONNX 106.501 ms 59 - 78 MB CPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 74.256 ms 12 - 28 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 108.806 ms 53 - 89 MB CPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 7.753 ms 0 - 41 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 SA7255P ADP Qualcomm® SA7255P QNN_DLC 10.569 ms 2 - 44 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 2.975 ms 0 - 195 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 4.632 ms 1 - 187 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 4.689 ms 0 - 47 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 SA8295P ADP Qualcomm® SA8295P QNN_DLC 6.588 ms 2 - 50 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 2.983 ms 0 - 193 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 4.63 ms 0 - 177 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 3.452 ms 0 - 41 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 SA8775P ADP Qualcomm® SA8775P QNN_DLC 5.118 ms 2 - 44 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 2.16 ms 8 - 87 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 3.304 ms 2 - 74 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 2.209 ms 0 - 78 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 1.799 ms 0 - 42 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 2.253 ms 2 - 51 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 1.833 ms 0 - 52 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 1.625 ms 0 - 43 MB NPU FFNet-78S-LowRes.tflite
FFNet-78S-LowRes w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 1.829 ms 2 - 50 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 1.61 ms 1 - 51 MB NPU FFNet-78S-LowRes.onnx.zip
FFNet-78S-LowRes w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 5.009 ms 179 - 179 MB NPU FFNet-78S-LowRes.dlc
FFNet-78S-LowRes w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 3.083 ms 25 - 25 MB NPU FFNet-78S-LowRes.onnx.zip

Installation

Install the package via pip:

# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[ffnet-78s-lowres]"

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.ffnet_78s_lowres.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.ffnet_78s_lowres.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.ffnet_78s_lowres.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.ffnet_78s_lowres 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. 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.ffnet_78s_lowres.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.ffnet_78s_lowres.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 FFNet-78S-LowRes's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of FFNet-78S-LowRes can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community

Downloads last month
249
Inference Providers NEW
Image Segmentation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support