Hugging Face's logo

qualcomm
/
FFNet-78S

Image Segmentation
PyTorch
android
Model card Files
xet
 Community

FFNet-78S: Optimized for Qualcomm Devices

FFNet-78S 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-78S 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-78S 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-78S on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Model checkpoint: ffnet78S_dBBB_cityscapes_state_dict_quarts
  • Input resolution: 2048x1024
  • Number of output classes: 19
  • Number of parameters: 27.5M
  • Model size (float): 105 MB
  • Model size (w8a8): 26.7 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
FFNet-78S ONNX float Snapdragon® X2 Elite 18.528 ms 24 - 24 MB NPU
FFNet-78S ONNX float Snapdragon® X Elite 37.713 ms 31 - 31 MB NPU
FFNet-78S ONNX float Snapdragon® 8 Gen 3 Mobile 27.589 ms 19 - 332 MB NPU
FFNet-78S ONNX float Snapdragon® 8 Gen 1 Mobile 59.731 ms 9 - 288 MB NPU
FFNet-78S ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 38.389 ms 24 - 395 MB NPU
FFNet-78S ONNX float Qualcomm® QCS8450 59.731 ms 9 - 288 MB NPU
FFNet-78S ONNX float Qualcomm® Dragonwing™ IQ-9075 56.807 ms 24 - 51 MB NPU
FFNet-78S ONNX float Qualcomm® Dragonwing™ IQ-X7181 37.713 ms 31 - 31 MB NPU
FFNet-78S ONNX float Qualcomm® Dragonwing™ Q-8750 21.667 ms 7 - 204 MB NPU
FFNet-78S ONNX float Snapdragon® 8 Elite Mobile 21.667 ms 7 - 204 MB NPU
FFNet-78S ONNX float Snapdragon® 8 Elite Gen 5 Mobile 17.722 ms 5 - 228 MB NPU
FFNet-78S ONNX w8a8 Snapdragon® X2 Elite 7.965 ms 7 - 7 MB NPU
FFNet-78S ONNX w8a8 Snapdragon® X Elite 13.545 ms 22 - 22 MB NPU
FFNet-78S ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 8.54 ms 5 - 297 MB NPU
FFNet-78S ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 17.858 ms 7 - 301 MB NPU
FFNet-78S ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 56.444 ms 6 - 9 MB NPU
FFNet-78S ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 12.808 ms 0 - 23 MB NPU
FFNet-78S ONNX w8a8 Qualcomm® QCS8450 17.858 ms 7 - 301 MB NPU
FFNet-78S ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 19.088 ms 6 - 9 MB NPU
FFNet-78S ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 13.545 ms 22 - 22 MB NPU
FFNet-78S ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 8.737 ms 1 - 215 MB NPU
FFNet-78S ONNX w8a8 Snapdragon® 8 Elite Mobile 8.737 ms 1 - 215 MB NPU
FFNet-78S ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 7.414 ms 2 - 219 MB NPU
FFNet-78S QNN_DLC float Snapdragon® X2 Elite 17.988 ms 24 - 24 MB NPU
FFNet-78S QNN_DLC float Snapdragon® X Elite 44.547 ms 24 - 24 MB NPU
FFNet-78S QNN_DLC float Snapdragon® 8 Gen 3 Mobile 29.441 ms 15 - 312 MB NPU
FFNet-78S QNN_DLC float Snapdragon® 8 Gen 1 Mobile 85.623 ms 6 - 290 MB NPU
FFNet-78S QNN_DLC float Qualcomm® Dragonwing™ QCS8275 188.034 ms 24 - 234 MB NPU
FFNet-78S QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 42.686 ms 24 - 26 MB NPU
FFNet-78S QNN_DLC float Qualcomm® SA8775P 60.885 ms 24 - 234 MB NPU
FFNet-78S QNN_DLC float Qualcomm® SA8650P 60.885 ms 24 - 234 MB NPU
FFNet-78S QNN_DLC float Qualcomm® SA8255P 60.885 ms 24 - 234 MB NPU
FFNet-78S QNN_DLC float Qualcomm® QCS8450 85.623 ms 6 - 290 MB NPU
FFNet-78S QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 66.625 ms 24 - 52 MB NPU
FFNet-78S QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 44.547 ms 24 - 24 MB NPU
FFNet-78S QNN_DLC float Qualcomm® Dragonwing™ Q-8750 21.86 ms 15 - 242 MB NPU
FFNet-78S QNN_DLC float Qualcomm® SA7255P 188.034 ms 24 - 234 MB NPU
FFNet-78S QNN_DLC float Qualcomm® SA8295P 66.414 ms 24 - 229 MB NPU
FFNet-78S QNN_DLC float Snapdragon® 8 Elite Mobile 21.86 ms 15 - 242 MB NPU
FFNet-78S QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 15.697 ms 14 - 266 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® X2 Elite 7.036 ms 6 - 6 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® X Elite 17.7 ms 6 - 6 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 11.63 ms 6 - 291 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 23.66 ms 6 - 293 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 64.262 ms 8 - 16 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8275 39.091 ms 6 - 225 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 16.828 ms 6 - 8 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® SA8775P 17.18 ms 6 - 225 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® SA8650P 17.18 ms 6 - 225 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® SA8255P 17.18 ms 6 - 225 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® QCS8450 23.66 ms 6 - 293 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 17.637 ms 6 - 14 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 17.7 ms 6 - 6 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 162.0 ms 6 - 254 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 22.675 ms 6 - 242 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 8.06 ms 6 - 238 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® SA7255P 39.091 ms 6 - 225 MB NPU
FFNet-78S QNN_DLC w8a8 Qualcomm® SA8295P 23.307 ms 6 - 227 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 8.06 ms 6 - 238 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 5.967 ms 6 - 251 MB NPU
FFNet-78S QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 22.675 ms 6 - 242 MB NPU
FFNet-78S TFLITE float Snapdragon® 8 Gen 3 Mobile 29.256 ms 1 - 377 MB NPU
FFNet-78S TFLITE float Snapdragon® 8 Gen 1 Mobile 86.333 ms 3 - 363 MB NPU
FFNet-78S TFLITE float Qualcomm® Dragonwing™ QCS8275 188.019 ms 3 - 245 MB NPU
FFNet-78S TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 43.073 ms 2 - 5 MB NPU
FFNet-78S TFLITE float Qualcomm® SA8775P 60.981 ms 3 - 246 MB NPU
FFNet-78S TFLITE float Qualcomm® SA8650P 60.981 ms 3 - 246 MB NPU
FFNet-78S TFLITE float Qualcomm® SA8255P 60.981 ms 3 - 246 MB NPU
FFNet-78S TFLITE float Qualcomm® QCS8450 86.333 ms 3 - 363 MB NPU
FFNet-78S TFLITE float Qualcomm® Dragonwing™ IQ-9075 61.731 ms 0 - 82 MB NPU
FFNet-78S TFLITE float Qualcomm® Dragonwing™ Q-8750 21.712 ms 2 - 263 MB NPU
FFNet-78S TFLITE float Qualcomm® SA7255P 188.019 ms 3 - 245 MB NPU
FFNet-78S TFLITE float Qualcomm® SA8295P 66.419 ms 3 - 240 MB NPU
FFNet-78S TFLITE float Snapdragon® 8 Elite Mobile 21.712 ms 2 - 263 MB NPU
FFNet-78S TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 15.458 ms 2 - 284 MB NPU
FFNet-78S TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 6.443 ms 1 - 288 MB NPU
FFNet-78S TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 18.231 ms 1 - 290 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 49.884 ms 1 - 36 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ QCS8275 26.23 ms 1 - 218 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.053 ms 1 - 3 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® SA8775P 9.747 ms 1 - 219 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® SA8650P 9.747 ms 1 - 219 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® SA8255P 9.747 ms 1 - 219 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® QCS8450 18.231 ms 1 - 290 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 9.478 ms 0 - 36 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 138.117 ms 0 - 246 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 14.527 ms 1 - 235 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 5.028 ms 1 - 231 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® SA7255P 26.23 ms 1 - 218 MB NPU
FFNet-78S TFLITE w8a8 Qualcomm® SA8295P 14.885 ms 1 - 223 MB NPU
FFNet-78S TFLITE w8a8 Snapdragon® 8 Elite Mobile 5.028 ms 1 - 231 MB NPU
FFNet-78S TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 3.657 ms 1 - 244 MB NPU
FFNet-78S TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 14.527 ms 1 - 235 MB NPU

License

  • The license for the original implementation of FFNet-78S 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-78S