PSPNet: Optimized for Qualcomm Devices

PSPNet (Pyramid Scene Parsing Network) is a semantic segmentation model that captures global context information by applying pyramid pooling modules. It is designed to improve scene understanding by aggregating contextual features at multiple scales.

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 PSPNet 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 PSPNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

Model checkpoint: pspnet101_ade20k.pth
Input resolution: 1x3x473x473
Number of parameters: 65.7M
Model size (float): 251 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
PSPNet	ONNX	float	Snapdragon® X2 Elite	428.799 ms	266 - 266 MB	NPU
PSPNet	ONNX	float	Snapdragon® X Elite	695.154 ms	529 - 529 MB	NPU
PSPNet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	513.442 ms	147 - 1066 MB	NPU
PSPNet	ONNX	float	Qualcomm® QCS8550 (Proxy)	606.249 ms	188 - 190 MB	NPU
PSPNet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	334.904 ms	248 - 1006 MB	NPU
PSPNet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	324.487 ms	224 - 992 MB	NPU
PSPNet	ONNX	w8a8	Snapdragon® X2 Elite	15.299 ms	131 - 131 MB	NPU
PSPNet	ONNX	w8a8	Snapdragon® X Elite	28.506 ms	133 - 133 MB	NPU
PSPNet	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	22.095 ms	71 - 492 MB	NPU
PSPNet	ONNX	w8a8	Qualcomm® QCS6490	3343.057 ms	204 - 285 MB	CPU
PSPNet	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	30.252 ms	32 - 115 MB	NPU
PSPNet	ONNX	w8a8	Qualcomm® QCS9075	34.05 ms	70 - 73 MB	NPU
PSPNet	ONNX	w8a8	Qualcomm® QCM6690	3145.346 ms	63 - 74 MB	CPU
PSPNet	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	19.001 ms	71 - 342 MB	NPU
PSPNet	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	2605.807 ms	91 - 103 MB	CPU
PSPNet	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	16.968 ms	71 - 340 MB	NPU
PSPNet	QNN_DLC	float	Snapdragon® X2 Elite	230.422 ms	3 - 3 MB	NPU
PSPNet	QNN_DLC	float	Snapdragon® X Elite	544.385 ms	3 - 3 MB	NPU
PSPNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	621.548 ms	0 - 873 MB	NPU
PSPNet	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	1865.275 ms	2 - 722 MB	NPU
PSPNet	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	609.258 ms	3 - 5 MB	NPU
PSPNet	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	1698.164 ms	0 - 429 MB	NPU
PSPNet	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	303.534 ms	141 - 860 MB	NPU
PSPNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	270.855 ms	1 - 734 MB	NPU
PSPNet	QNN_DLC	w8a8	Snapdragon® X2 Elite	14.594 ms	1 - 1 MB	NPU
PSPNet	QNN_DLC	w8a8	Snapdragon® X Elite	26.084 ms	1 - 1 MB	NPU
PSPNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	21.318 ms	1 - 362 MB	NPU
PSPNet	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	63.872 ms	1 - 263 MB	NPU
PSPNet	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	27.353 ms	1 - 3 MB	NPU
PSPNet	QNN_DLC	w8a8	Qualcomm® QCS9075	32.019 ms	1 - 35 MB	NPU
PSPNet	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	38.613 ms	1 - 360 MB	NPU
PSPNet	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	18.41 ms	1 - 233 MB	NPU
PSPNet	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	91.51 ms	1 - 430 MB	NPU
PSPNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	15.938 ms	1 - 238 MB	NPU
PSPNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	530.504 ms	120 - 1170 MB	NPU
PSPNet	TFLITE	float	Qualcomm® QCS8275 (Proxy)	1684.384 ms	121 - 956 MB	NPU
PSPNet	TFLITE	float	Qualcomm® QCS8550 (Proxy)	617.858 ms	121 - 124 MB	NPU
PSPNet	TFLITE	float	Qualcomm® QCS8450 (Proxy)	1564.358 ms	44 - 617 MB	NPU
PSPNet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	971.866 ms	38 - 795 MB	NPU
PSPNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	328.449 ms	128 - 965 MB	NPU
PSPNet	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	21.514 ms	50 - 437 MB	NPU
PSPNet	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	66.574 ms	32 - 304 MB	NPU
PSPNet	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	30.01 ms	32 - 34 MB	NPU
PSPNet	TFLITE	w8a8	Qualcomm® QCS9075	32.442 ms	32 - 131 MB	NPU
PSPNet	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	40.322 ms	18 - 403 MB	NPU
PSPNet	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	18.543 ms	32 - 285 MB	NPU
PSPNet	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	95.006 ms	32 - 445 MB	NPU
PSPNet	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	15.312 ms	32 - 283 MB	NPU

License

The license for the original implementation of PSPNet can be found here.

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.

Downloads last month: -; Downloads are not tracked for this model. How to track