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® X Elite	9.104 ms	56 - 56 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® 8 Gen 3 Mobile	8.026 ms	7 - 226 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Qualcomm® QCS8550 (Proxy)	9.826 ms	0 - 60 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Qualcomm® QCS9075	10.224 ms	6 - 15 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	6.378 ms	2 - 181 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.491 ms	2 - 181 MB	NPU
FFNet-122NS-LowRes	ONNX	float	Snapdragon® X2 Elite	3.611 ms	57 - 57 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® X Elite	2.825 ms	30 - 30 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	1.886 ms	0 - 153 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCS6490	91.515 ms	54 - 150 MB	CPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	2.614 ms	0 - 33 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCS9075	3.13 ms	1 - 4 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Qualcomm® QCM6690	85.533 ms	58 - 71 MB	CPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.466 ms	0 - 93 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	85.114 ms	59 - 72 MB	CPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.323 ms	0 - 90 MB	NPU
FFNet-122NS-LowRes	ONNX	w8a8	Snapdragon® X2 Elite	1.271 ms	30 - 30 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® X Elite	13.0 ms	6 - 6 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	8.774 ms	6 - 207 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	39.551 ms	0 - 161 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	12.839 ms	6 - 8 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® SA8775P	16.029 ms	1 - 162 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS9075	16.713 ms	6 - 14 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	29.898 ms	6 - 195 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® SA7255P	39.551 ms	0 - 161 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Qualcomm® SA8295P	17.709 ms	0 - 157 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	6.971 ms	6 - 169 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	5.635 ms	6 - 172 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	float	Snapdragon® X2 Elite	6.342 ms	6 - 6 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® X Elite	4.834 ms	2 - 2 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	3.175 ms	2 - 131 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS6490	13.818 ms	2 - 5 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	9.28 ms	1 - 76 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	4.511 ms	0 - 6 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® SA8775P	21.251 ms	1 - 76 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS9075	5.862 ms	1 - 5 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCM6690	25.715 ms	2 - 200 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	6.856 ms	2 - 127 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® SA7255P	9.28 ms	1 - 76 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Qualcomm® SA8295P	6.044 ms	1 - 74 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	2.183 ms	2 - 79 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	5.745 ms	2 - 195 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.798 ms	2 - 79 MB	NPU
FFNet-122NS-LowRes	QNN_DLC	w8a8	Snapdragon® X2 Elite	2.031 ms	2 - 2 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	8.62 ms	0 - 277 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS8275 (Proxy)	39.9 ms	1 - 193 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS8550 (Proxy)	12.787 ms	1 - 3 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® SA8775P	15.996 ms	1 - 193 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS9075	16.022 ms	0 - 70 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® QCS8450 (Proxy)	29.684 ms	1 - 267 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® SA7255P	39.9 ms	1 - 193 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Qualcomm® SA8295P	18.025 ms	1 - 191 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	6.866 ms	1 - 193 MB	NPU
FFNet-122NS-LowRes	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	5.428 ms	1 - 198 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	1.863 ms	0 - 134 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS6490	8.834 ms	0 - 35 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	5.97 ms	0 - 72 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	2.597 ms	0 - 3 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® SA8775P	3.04 ms	0 - 74 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS9075	3.072 ms	0 - 35 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCM6690	22.072 ms	0 - 197 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	3.283 ms	0 - 130 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® SA7255P	5.97 ms	0 - 72 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Qualcomm® SA8295P	3.846 ms	0 - 69 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.439 ms	0 - 77 MB	NPU
FFNet-122NS-LowRes	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.72 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.