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 14.263 ms 24 - 24 MB NPU
FFNet-40S ONNX float Snapdragon® X Elite 31.29 ms 24 - 24 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Gen 3 Mobile 22.611 ms 29 - 318 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Gen 1 Mobile 52.446 ms 0 - 259 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ IQ-8275 43.8 ms 24 - 28 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 31.454 ms 25 - 55 MB NPU
FFNet-40S ONNX float Qualcomm® QCS8450 52.446 ms 0 - 259 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ IQ-X7181 31.29 ms 24 - 24 MB NPU
FFNet-40S ONNX float Qualcomm® Dragonwing™ Q-8750 17.55 ms 7 - 192 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Elite Mobile 17.55 ms 7 - 192 MB NPU
FFNet-40S ONNX float Snapdragon® 8 Elite Gen 5 Mobile 13.501 ms 5 - 217 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® X2 Elite 6.863 ms 7 - 7 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® X Elite 11.763 ms 8 - 8 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 7.387 ms 7 - 253 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 15.768 ms 7 - 255 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 49.031 ms 6 - 9 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 10.829 ms 6 - 10 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 11.245 ms 0 - 177 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® QCS8450 15.768 ms 7 - 255 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 11.911 ms 6 - 9 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 11.763 ms 8 - 8 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 112.29 ms 5 - 242 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 14.817 ms 7 - 223 MB NPU
FFNet-40S ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 7.67 ms 1 - 190 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Elite Mobile 7.67 ms 1 - 190 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 6.63 ms 1 - 194 MB NPU
FFNet-40S ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 14.817 ms 7 - 223 MB NPU
FFNet-40S QNN_DLC float Snapdragon® X Elite 37.992 ms 24 - 24 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Gen 3 Mobile 24.747 ms 43 - 313 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Gen 1 Mobile 63.476 ms 24 - 289 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 48.627 ms 24 - 53 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 135.567 ms 24 - 218 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 36.75 ms 24 - 26 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8775P 48.919 ms 24 - 219 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8650P 48.919 ms 24 - 219 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8255P 48.919 ms 24 - 219 MB NPU
FFNet-40S QNN_DLC float Qualcomm® QCS8450 63.476 ms 24 - 289 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 49.736 ms 24 - 52 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 37.992 ms 24 - 24 MB NPU
FFNet-40S QNN_DLC float Qualcomm® Dragonwing™ Q-8750 18.183 ms 23 - 246 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA7255P 135.567 ms 24 - 218 MB NPU
FFNet-40S QNN_DLC float Qualcomm® SA8295P 53.706 ms 24 - 223 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Elite Mobile 18.183 ms 23 - 246 MB NPU
FFNet-40S QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 12.594 ms 16 - 257 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® X2 Elite 6.092 ms 6 - 6 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® X Elite 16.041 ms 6 - 6 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 10.487 ms 6 - 252 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 21.261 ms 6 - 254 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 55.894 ms 6 - 14 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 14.787 ms 6 - 15 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 32.925 ms 6 - 202 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 15.077 ms 6 - 8 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8775P 15.686 ms 6 - 203 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8650P 15.686 ms 6 - 203 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8255P 15.686 ms 6 - 203 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® QCS8450 21.261 ms 6 - 254 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 17.719 ms 4 - 12 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 16.041 ms 6 - 6 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 125.501 ms 6 - 238 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 19.036 ms 6 - 223 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 7.062 ms 6 - 218 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA7255P 32.925 ms 6 - 202 MB NPU
FFNet-40S QNN_DLC w8a8 Qualcomm® SA8295P 20.056 ms 6 - 206 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 7.062 ms 6 - 218 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 5.294 ms 6 - 233 MB NPU
FFNet-40S QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 19.036 ms 6 - 223 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Gen 3 Mobile 25.202 ms 1 - 306 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Gen 1 Mobile 63.403 ms 3 - 304 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ IQ-8275 48.268 ms 0 - 57 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ IQ-8275 135.663 ms 3 - 210 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 36.171 ms 2 - 5 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8775P 49.011 ms 2 - 211 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8650P 49.011 ms 2 - 211 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8255P 49.011 ms 2 - 211 MB NPU
FFNet-40S TFLITE float Qualcomm® QCS8450 63.403 ms 3 - 304 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ IQ-9075 49.975 ms 0 - 56 MB NPU
FFNet-40S TFLITE float Qualcomm® Dragonwing™ Q-8750 18.007 ms 0 - 236 MB NPU
FFNet-40S TFLITE float Qualcomm® SA7255P 135.663 ms 3 - 210 MB NPU
FFNet-40S TFLITE float Qualcomm® SA8295P 53.699 ms 2 - 215 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Elite Mobile 18.007 ms 0 - 236 MB NPU
FFNet-40S TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 12.814 ms 2 - 253 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 5.465 ms 1 - 245 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 12.362 ms 1 - 248 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 40.63 ms 0 - 23 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 8.012 ms 1 - 24 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 20.074 ms 1 - 196 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 7.474 ms 1 - 3 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8775P 8.251 ms 1 - 197 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8650P 8.251 ms 1 - 197 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8255P 8.251 ms 1 - 197 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® QCS8450 12.362 ms 1 - 248 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 8.044 ms 0 - 23 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 104.497 ms 1 - 231 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 11.468 ms 1 - 216 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 3.977 ms 1 - 209 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA7255P 20.074 ms 1 - 196 MB NPU
FFNet-40S TFLITE w8a8 Qualcomm® SA8295P 11.683 ms 1 - 199 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Elite Mobile 3.977 ms 1 - 209 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 2.976 ms 1 - 226 MB NPU
FFNet-40S TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 11.468 ms 1 - 216 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