Hugging Face's logo

qualcomm
/
FFNet-40S

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
 Community

FFNet-40S: Optimized for Qualcomm Devices

FFNet-40S 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 is based on the implementation of FFNet-40S found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime Precision Chipset SDK Versions Download
ONNX float Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
QNN_DLC float Universal QAIRT 2.45 Download
QNN_DLC w8a8 Universal QAIRT 2.45 Download
TFLITE float Universal QAIRT 2.45 Download
TFLITE w8a8 Universal QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit FFNet-40S on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

  • Custom weights (e.g., fine-tuned checkpoints)
  • Custom input shapes
  • Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for FFNet-40S on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Input resolution: 2048x1024
  • Model checkpoint: ffnet40S_dBBB_cityscapes_state_dict_quarts
  • Model size (float): 53.1 MB
  • Model size (w8a8): 13.5 MB
  • Number of output classes: 19
  • Number of parameters: 13.9M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
FFNet-40S ONNX float Snapdragon® X2 Elite 14.23 ms 24 - 24 MB NPU
FFNet-40S ONNX float Snapdragon® X Elite 31.281 ms 24 - 24 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Gen 3 Mobile 22.284 ms 28 - 315 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Gen 1 Mobile 52.151 ms 25 - 283 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ IQ-8275 43.514 ms 24 - 28 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 32.036 ms 24 - 54 MB NPU
FFNet-40S ONNX float Qualcomm® QCS8450 52.151 ms 25 - 283 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ IQ-9075 44.163 ms 24 - 27 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ IQ-X7181 31.281 ms 24 - 24 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ Q-8750 17.857 ms 6 - 191 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Elite Mobile 17.857 ms 6 - 191 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Elite Gen 5 Mobile 13.5 ms 5 - 217 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® X2 Elite 6.917 ms 7 - 7 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® X Elite 11.817 ms 9 - 9 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 7.381 ms 7 - 255 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 15.776 ms 7 - 254 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 49.013 ms 6 - 9 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 10.863 ms 6 - 10 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 11.225 ms 0 - 14 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® QCS8450 15.776 ms 7 - 254 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 12.223 ms 6 - 9 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 11.817 ms 9 - 9 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 111.413 ms 5 - 242 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 14.82 ms 7 - 223 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 7.649 ms 1 - 190 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Elite Mobile 7.649 ms 1 - 190 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 6.625 ms 1 - 195 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 14.82 ms 7 - 223 MB NPU
FFNet-40S QNN_DLC float Snapdragon® X2 Elite 14.533 ms 24 - 24 MB NPU
FFNet-40S QNN_DLC float Snapdragon® X Elite 37.477 ms 24 - 24 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Gen 3 Mobile 24.751 ms 22 - 290 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Gen 1 Mobile 63.859 ms 24 - 289 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 48.234 ms 24 - 53 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 135.507 ms 24 - 218 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 36.443 ms 24 - 38 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8775P 48.952 ms 24 - 219 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8650P 48.952 ms 24 - 219 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8255P 48.952 ms 24 - 219 MB NPU
FFNet-40S QNN_DLC float Qualcomm® QCS8450 63.859 ms 24 - 289 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 54.076 ms 24 - 52 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 37.477 ms 24 - 24 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ Q-8750 18.159 ms 18 - 241 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA7255P 135.507 ms 24 - 218 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8295P 53.704 ms 24 - 225 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Elite Mobile 18.159 ms 18 - 241 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 12.712 ms 22 - 264 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® X2 Elite 6.103 ms 6 - 6 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® X Elite 16.053 ms 6 - 6 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 10.508 ms 6 - 252 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 21.258 ms 6 - 254 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 55.551 ms 6 - 14 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 14.905 ms 6 - 15 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 32.897 ms 6 - 202 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 15.035 ms 6 - 8 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8775P 15.688 ms 6 - 203 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8650P 15.688 ms 6 - 203 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8255P 15.688 ms 6 - 203 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® QCS8450 21.258 ms 6 - 254 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 15.686 ms 6 - 14 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 16.053 ms 6 - 6 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 128.325 ms 7 - 239 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 19.17 ms 6 - 225 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 7.075 ms 6 - 216 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA7255P 32.897 ms 6 - 202 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8295P 20.061 ms 6 - 206 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 7.075 ms 6 - 216 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 5.292 ms 6 - 232 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 19.17 ms 6 - 225 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Gen 3 Mobile 24.954 ms 1 - 304 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Gen 1 Mobile 63.435 ms 3 - 304 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ IQ-8275 48.267 ms 0 - 57 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ IQ-8275 135.586 ms 0 - 208 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 36.54 ms 3 - 5 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8775P 48.984 ms 0 - 208 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8650P 48.984 ms 0 - 208 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8255P 48.984 ms 0 - 208 MB NPU
FFNet-40S TFLITE float Qualcomm® QCS8450 63.435 ms 3 - 304 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ IQ-9075 49.9 ms 0 - 56 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ Q-8750 18.038 ms 0 - 235 MB NPU
FFNet-40S TFLITE float Qualcomm® SA7255P 135.586 ms 0 - 208 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8295P 53.712 ms 2 - 215 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Elite Mobile 18.038 ms 0 - 235 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 12.764 ms 2 - 253 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 5.459 ms 1 - 249 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 12.349 ms 0 - 248 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 41.165 ms 0 - 23 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 8.019 ms 1 - 24 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 20.072 ms 1 - 196 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 7.457 ms 1 - 3 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8775P 8.254 ms 1 - 197 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8650P 8.254 ms 1 - 197 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8255P 8.254 ms 1 - 197 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® QCS8450 12.349 ms 0 - 248 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 8.061 ms 1 - 23 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 102.085 ms 3 - 232 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 11.481 ms 1 - 217 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 3.988 ms 1 - 212 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA7255P 20.072 ms 1 - 196 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8295P 11.699 ms 1 - 201 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Elite Mobile 3.988 ms 1 - 212 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 2.967 ms 1 - 225 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 11.481 ms 1 - 217 MB NPU

License

  • The license for the original implementation of FFNet-40S can be found here.

References

Community

Downloads last month

-

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

Paper for qualcomm/FFNet-40S