---
library_name: pytorch
license: other
tags:
- android
pipeline_tag: image-segmentation

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/web-assets/model_demo.png)

# FFNet-122NS-LowRes: Optimized for Qualcomm Devices

FFNet-122NS-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 is based on the implementation of FFNet-122NS-LowRes found [here](https://github.com/Qualcomm-AI-research/FFNet).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/ffnet_122ns_lowres) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/releases/v0.51.0/ffnet_122ns_lowres-onnx-float.zip)
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/releases/v0.51.0/ffnet_122ns_lowres-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/releases/v0.51.0/ffnet_122ns_lowres-qnn_dlc-float.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/releases/v0.51.0/ffnet_122ns_lowres-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/releases/v0.51.0/ffnet_122ns_lowres-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ffnet_122ns_lowres/releases/v0.51.0/ffnet_122ns_lowres-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[FFNet-122NS-LowRes on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/ffnet_122ns_lowres)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/ffnet_122ns_lowres) 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-122NS-LowRes on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/ffnet_122ns_lowres) for usage instructions.

## Model Details

**Model Type:** Model_use_case.semantic_segmentation

**Model Stats:**
- Model checkpoint: ffnet122NS_CCC_cityscapes_state_dict_quarts_pre_down
- Input resolution: 1024x512
- Number of output classes: 19
- Number of parameters: 32.1M
- Model size (float): 123 MB
- Model size (w8a8): 31.3 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| FFNet-122NS-LowRes | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.497 ms | 0 - 179 MB | NPU
| FFNet-122NS-LowRes | ONNX | float | Snapdragon® X2 Elite | 3.583 ms | 57 - 57 MB | NPU
| FFNet-122NS-LowRes | ONNX | float | Snapdragon® X Elite | 9.055 ms | 56 - 56 MB | NPU
| FFNet-122NS-LowRes | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 7.969 ms | 7 - 228 MB | NPU
| FFNet-122NS-LowRes | ONNX | float | Qualcomm® QCS8550 (Proxy) | 9.905 ms | 0 - 60 MB | NPU
| FFNet-122NS-LowRes | ONNX | float | Qualcomm® QCS9075 | 10.219 ms | 6 - 15 MB | NPU
| FFNet-122NS-LowRes | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.383 ms | 2 - 172 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.323 ms | 0 - 87 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Snapdragon® X2 Elite | 1.283 ms | 30 - 30 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Snapdragon® X Elite | 2.803 ms | 30 - 30 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.884 ms | 0 - 153 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Qualcomm® QCS6490 | 89.666 ms | 54 - 151 MB | CPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 2.62 ms | 0 - 109 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Qualcomm® QCS9075 | 3.131 ms | 1 - 4 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Qualcomm® QCM6690 | 85.544 ms | 50 - 64 MB | CPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 1.467 ms | 0 - 92 MB | NPU
| FFNet-122NS-LowRes | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 85.263 ms | 61 - 75 MB | CPU
| FFNet-122NS-LowRes | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.601 ms | 6 - 172 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Snapdragon® X2 Elite | 6.36 ms | 6 - 6 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Snapdragon® X Elite | 13.3 ms | 6 - 6 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.797 ms | 6 - 200 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 39.409 ms | 1 - 160 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 13.042 ms | 6 - 8 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® SA8775P | 16.005 ms | 0 - 161 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® QCS9075 | 16.656 ms | 6 - 14 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 30.378 ms | 6 - 193 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® SA7255P | 39.409 ms | 1 - 160 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Qualcomm® SA8295P | 18.062 ms | 0 - 153 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.915 ms | 6 - 168 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.922 ms | 2 - 81 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 2.135 ms | 2 - 2 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Snapdragon® X Elite | 4.934 ms | 2 - 2 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 3.232 ms | 2 - 130 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 14.112 ms | 3 - 7 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 9.412 ms | 2 - 77 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 4.58 ms | 2 - 3 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® SA8775P | 5.093 ms | 2 - 80 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 6.078 ms | 1 - 4 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 28.908 ms | 2 - 203 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 6.621 ms | 2 - 128 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® SA7255P | 9.412 ms | 2 - 77 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Qualcomm® SA8295P | 6.144 ms | 1 - 76 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 2.221 ms | 2 - 78 MB | NPU
| FFNet-122NS-LowRes | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 5.786 ms | 2 - 194 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.348 ms | 0 - 197 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.616 ms | 0 - 273 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 39.982 ms | 1 - 190 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 12.865 ms | 1 - 22 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® SA8775P | 16.165 ms | 1 - 191 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® QCS9075 | 16.851 ms | 0 - 71 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 29.74 ms | 1 - 267 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® SA7255P | 39.982 ms | 1 - 190 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Qualcomm® SA8295P | 17.911 ms | 1 - 187 MB | NPU
| FFNet-122NS-LowRes | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.829 ms | 0 - 196 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.318 ms | 0 - 75 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.893 ms | 0 - 133 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® QCS6490 | 9.272 ms | 0 - 35 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 6.07 ms | 0 - 72 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 2.707 ms | 0 - 7 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® SA8775P | 3.117 ms | 0 - 74 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® QCS9075 | 3.118 ms | 0 - 35 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® QCM6690 | 22.643 ms | 0 - 195 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 3.341 ms | 0 - 128 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® SA7255P | 6.07 ms | 0 - 72 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Qualcomm® SA8295P | 3.945 ms | 0 - 70 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 1.454 ms | 0 - 71 MB | NPU
| FFNet-122NS-LowRes | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 3.989 ms | 0 - 186 MB | NPU

## License
* The license for the original implementation of FFNet-122NS-LowRes can be found
  [here](https://github.com/Qualcomm-AI-research/FFNet/blob/master/LICENSE).

## References
* [Simple and Efficient Architectures for Semantic Segmentation](https://arxiv.org/abs/2206.08236)
* [Source Model Implementation](https://github.com/Qualcomm-AI-research/FFNet)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).