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. 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 1.24.1	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.17.0	Download

For more device-specific assets and performance metrics, visit FFNet-122NS-LowRes 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-122NS-LowRes on GitHub 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® X2 Elite	3.603 ms	57 - 57 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® X Elite	9.117 ms	56 - 56 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® 8 Gen 3 Mobile	7.998 ms	7 - 227 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Qualcomm® QCS8550 (Proxy)	9.809 ms	0 - 60 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Qualcomm® QCS9075	10.21 ms	6 - 15 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	6.362 ms	2 - 180 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.494 ms	2 - 181 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® X2 Elite	1.287 ms	30 - 30 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® X Elite	2.816 ms	30 - 30 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	1.892 ms	0 - 156 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCS6490	89.433 ms	54 - 151 MB	CPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	2.615 ms	0 - 33 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCS9075	3.122 ms	1 - 4 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCM6690	85.97 ms	58 - 72 MB	CPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.465 ms	0 - 85 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	85.388 ms	59 - 71 MB	CPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.322 ms	0 - 91 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® X2 Elite	6.419 ms	6 - 6 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® X Elite	13.003 ms	6 - 6 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	8.759 ms	6 - 205 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	39.592 ms	1 - 161 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	12.865 ms	6 - 13 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® SA8775P	15.987 ms	0 - 161 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS9075	16.521 ms	6 - 14 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	30.074 ms	6 - 195 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® SA7255P	39.592 ms	1 - 161 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® SA8295P	17.704 ms	0 - 158 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	6.989 ms	0 - 163 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	5.667 ms	6 - 172 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® X2 Elite	2.042 ms	2 - 2 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® X Elite	4.847 ms	2 - 2 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	3.178 ms	2 - 131 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS6490	13.874 ms	1 - 5 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	9.241 ms	1 - 76 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	4.51 ms	2 - 3 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® SA8775P	4.982 ms	2 - 79 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS9075	5.872 ms	3 - 6 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCM6690	27.567 ms	2 - 201 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	6.902 ms	2 - 126 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® SA7255P	9.241 ms	1 - 76 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® SA8295P	6.042 ms	1 - 74 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	2.181 ms	2 - 75 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	5.735 ms	2 - 193 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.799 ms	2 - 81 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	8.598 ms	0 - 277 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS8275 (Proxy)	39.971 ms	1 - 193 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS8550 (Proxy)	12.739 ms	1 - 3 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® SA8775P	71.437 ms	1 - 192 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS9075	16.693 ms	0 - 70 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS8450 (Proxy)	30.128 ms	1 - 271 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® SA7255P	39.971 ms	1 - 193 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® SA8295P	18.009 ms	1 - 189 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	6.861 ms	1 - 199 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	5.394 ms	1 - 201 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	1.862 ms	0 - 128 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS6490	9.205 ms	0 - 35 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	5.932 ms	0 - 72 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	2.597 ms	0 - 4 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® SA8775P	3.051 ms	0 - 74 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS9075	3.069 ms	0 - 35 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCM6690	23.155 ms	0 - 197 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	3.249 ms	0 - 135 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® SA7255P	5.932 ms	0 - 72 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® SA8295P	3.879 ms	0 - 69 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.44 ms	0 - 76 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.708 ms	0 - 187 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.313 ms	0 - 75 MB	NPU

License

• The license for the original implementation of FFNet-122NS-LowRes can be found here.

References

• Simple and Efficient Architectures for Semantic Segmentation
• Source Model Implementation

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.