Hugging Face's logo

qualcomm
/
FFNet-54S

Image Segmentation
PyTorch
android
Model card Files
xet
 Community

FFNet-54S: Optimized for Qualcomm Devices

FFNet-54S 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-54S 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 2.42, ONNX Runtime 1.24.3 Download
ONNX w8a8 Universal QAIRT 2.42, ONNX Runtime 1.24.3 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-54S 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-54S on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Model checkpoint: ffnet54S_dBBB_cityscapes_state_dict_quarts
  • Input resolution: 2048x1024
  • Number of output classes: 19
  • Number of parameters: 18.0M
  • Model size (float): 68.8 MB
  • Model size (w8a8): 17.5 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
FFNet-54S ONNX float Snapdragon® 8 Elite Gen 5 Mobile 13.507 ms 13 - 240 MB NPU
FFNet-54S ONNX float Snapdragon® 8 Elite Mobile 17.869 ms 5 - 202 MB NPU
FFNet-54S ONNX float Snapdragon® X2 Elite 14.74 ms 22 - 22 MB NPU
FFNet-54S ONNX float Snapdragon® X Elite 34.042 ms 24 - 24 MB NPU
FFNet-54S ONNX float Snapdragon® X Elite 34.042 ms 24 - 24 MB NPU
FFNet-54S ONNX float Snapdragon® 8 Gen 3 Mobile 24.304 ms 29 - 314 MB NPU
FFNet-54S ONNX float Qualcomm® QCS8550 (Proxy) 34.802 ms 24 - 27 MB NPU
FFNet-54S ONNX float Snapdragon® 8 Elite For Galaxy Mobile 17.869 ms 5 - 202 MB NPU
FFNet-54S ONNX float Qualcomm® QCS9075 52.903 ms 24 - 51 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 6.915 ms 6 - 209 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® 8 Elite Mobile 11.634 ms 1 - 200 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® X2 Elite 7.396 ms 13 - 13 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® X Elite 11.193 ms 12 - 12 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® X Elite 11.193 ms 12 - 12 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 6.957 ms 7 - 264 MB NPU
FFNet-54S ONNX w8a8 Qualcomm® QCS6490 408.146 ms 182 - 237 MB CPU
FFNet-54S ONNX w8a8 Qualcomm® QCS8550 (Proxy) 10.564 ms 6 - 46 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 429.692 ms 156 - 167 MB CPU
FFNet-54S ONNX w8a8 Qualcomm® QCM6690 433.377 ms 146 - 156 MB CPU
FFNet-54S ONNX w8a8 Qualcomm® QCS9075 12.873 ms 6 - 9 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 11.634 ms 1 - 200 MB NPU
FFNet-54S ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 429.692 ms 156 - 167 MB CPU
FFNet-54S QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 13.695 ms 14 - 256 MB NPU
FFNet-54S QNN_DLC float Snapdragon® 8 Elite Mobile 19.66 ms 16 - 237 MB NPU
FFNet-54S QNN_DLC float Snapdragon® X2 Elite 15.987 ms 24 - 24 MB NPU
FFNet-54S QNN_DLC float Snapdragon® X Elite 39.8 ms 24 - 24 MB NPU
FFNet-54S QNN_DLC float Snapdragon® X Elite 39.8 ms 24 - 24 MB NPU
FFNet-54S QNN_DLC float Snapdragon® 8 Gen 3 Mobile 26.841 ms 9 - 290 MB NPU
FFNet-54S QNN_DLC float Qualcomm® QCS8550 (Proxy) 38.393 ms 24 - 45 MB NPU
FFNet-54S QNN_DLC float Qualcomm® SA8775P 53.67 ms 24 - 224 MB NPU
FFNet-54S QNN_DLC float Qualcomm® SA8775P 53.67 ms 24 - 224 MB NPU
FFNet-54S QNN_DLC float Qualcomm® SA8775P 53.67 ms 24 - 224 MB NPU
FFNet-54S QNN_DLC float Qualcomm® SA7255P 155.275 ms 16 - 216 MB NPU
FFNet-54S QNN_DLC float Qualcomm® QCS8450 (Proxy) 76.946 ms 24 - 296 MB NPU
FFNet-54S QNN_DLC float Qualcomm® SA8295P 59.245 ms 24 - 223 MB NPU
FFNet-54S QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 19.66 ms 16 - 237 MB NPU
FFNet-54S QNN_DLC float Qualcomm® QCS9075 66.829 ms 24 - 52 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 5.495 ms 6 - 239 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 7.48 ms 6 - 225 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® X2 Elite 6.473 ms 6 - 6 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® X Elite 16.559 ms 6 - 6 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® X Elite 16.559 ms 6 - 6 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 10.937 ms 6 - 264 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® QCS6490 72.079 ms 6 - 14 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 15.646 ms 6 - 15 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® SA8775P 16.294 ms 6 - 212 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® SA8775P 16.294 ms 6 - 212 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® SA8775P 16.294 ms 6 - 212 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 20.346 ms 6 - 230 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® QCM6690 141.519 ms 6 - 243 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® QCS9075 18.91 ms 6 - 14 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 7.48 ms 6 - 225 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® QCS8450 (Proxy) 22.377 ms 6 - 266 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® SA7255P 35.577 ms 6 - 211 MB NPU
FFNet-54S QNN_DLC w8a8 Qualcomm® SA8295P 21.311 ms 6 - 215 MB NPU
FFNet-54S QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 20.346 ms 6 - 230 MB NPU
FFNet-54S TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 13.686 ms 2 - 261 MB NPU
FFNet-54S TFLITE float Snapdragon® 8 Elite Mobile 19.596 ms 1 - 241 MB NPU
FFNet-54S TFLITE float Snapdragon® 8 Gen 3 Mobile 26.467 ms 2 - 332 MB NPU
FFNet-54S TFLITE float Qualcomm® QCS8550 (Proxy) 38.564 ms 2 - 5 MB NPU
FFNet-54S TFLITE float Qualcomm® SA8775P 53.71 ms 0 - 221 MB NPU
FFNet-54S TFLITE float Qualcomm® SA8775P 53.71 ms 0 - 221 MB NPU
FFNet-54S TFLITE float Qualcomm® SA8775P 53.71 ms 0 - 221 MB NPU
FFNet-54S TFLITE float Qualcomm® SA7255P 155.448 ms 3 - 224 MB NPU
FFNet-54S TFLITE float Qualcomm® QCS8450 (Proxy) 77.854 ms 2 - 321 MB NPU
FFNet-54S TFLITE float Qualcomm® SA8295P 59.282 ms 0 - 216 MB NPU
FFNet-54S TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 19.596 ms 1 - 241 MB NPU
FFNet-54S TFLITE float Qualcomm® QCS9075 67.401 ms 0 - 64 MB NPU
FFNet-54S TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 3.17 ms 1 - 232 MB NPU
FFNet-54S TFLITE w8a8 Snapdragon® 8 Elite Mobile 4.462 ms 0 - 221 MB NPU
FFNet-54S TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 5.915 ms 1 - 260 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® QCS6490 55.691 ms 0 - 27 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 8.29 ms 1 - 24 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® SA8775P 8.869 ms 1 - 205 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® SA8775P 8.869 ms 1 - 205 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® SA8775P 8.869 ms 1 - 205 MB NPU
FFNet-54S TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 12.596 ms 0 - 221 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® QCM6690 120.032 ms 1 - 238 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® QCS9075 10.055 ms 1 - 27 MB NPU
FFNet-54S TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 4.462 ms 0 - 221 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® QCS8450 (Proxy) 13.27 ms 1 - 263 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® SA7255P 22.855 ms 1 - 203 MB NPU
FFNet-54S TFLITE w8a8 Qualcomm® SA8295P 13.181 ms 1 - 207 MB NPU
FFNet-54S TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 12.596 ms 0 - 221 MB NPU

License

  • The license for the original implementation of FFNet-54S 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-54S