DnCNN: Optimized for Qualcomm Devices

DnCNN is a 17-layer denoising convolutional neural network that uses residual learning to remove Gaussian noise (sigma=25) from grayscale images. The network predicts the noise residual and subtracts it from the input to produce a clean image.

This is based on the implementation of DnCNN 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 1.24.3	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	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 DnCNN 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 DnCNN on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_editing

Model Stats:

Model checkpoint: dncnn_25
Input resolution: 256x256
Number of parameters: 555K
Model size (float): 2.12 MB
Model size (w8a8): 581 KB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
DnCNN	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.119 ms	0 - 142 MB	NPU
DnCNN	ONNX	float	Snapdragon® 8 Elite Mobile	4.139 ms	0 - 142 MB	NPU
DnCNN	ONNX	float	Snapdragon® X2 Elite	4.045 ms	0 - 0 MB	NPU
DnCNN	ONNX	float	Snapdragon® X Elite	7.396 ms	0 - 0 MB	NPU
DnCNN	ONNX	float	Snapdragon® X Elite	7.396 ms	0 - 0 MB	NPU
DnCNN	ONNX	float	Snapdragon® 8 Gen 3 Mobile	5.186 ms	1 - 181 MB	NPU
DnCNN	ONNX	float	Qualcomm® QCS8550 (Proxy)	6.928 ms	7 - 9 MB	NPU
DnCNN	ONNX	float	Qualcomm® QCS9075	14.572 ms	1 - 4 MB	NPU
DnCNN	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	4.139 ms	0 - 142 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.755 ms	0 - 33 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® 8 Elite Mobile	1.219 ms	0 - 34 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® X2 Elite	1.047 ms	0 - 0 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® X Elite	2.021 ms	0 - 0 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® X Elite	2.021 ms	0 - 0 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	1.338 ms	0 - 48 MB	NPU
DnCNN	ONNX	w8a8	Qualcomm® QCS6490	132.205 ms	60 - 63 MB	CPU
DnCNN	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	1.79 ms	0 - 3 MB	NPU
DnCNN	ONNX	w8a8	Qualcomm® QCS9075	1.933 ms	0 - 3 MB	NPU
DnCNN	ONNX	w8a8	Qualcomm® QCM6690	192.093 ms	61 - 68 MB	CPU
DnCNN	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.219 ms	0 - 34 MB	NPU
DnCNN	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	160.759 ms	62 - 69 MB	CPU
DnCNN	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	160.759 ms	62 - 69 MB	CPU
DnCNN	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	2.962 ms	0 - 146 MB	NPU
DnCNN	QNN_DLC	float	Snapdragon® 8 Elite Mobile	4.029 ms	0 - 142 MB	NPU
DnCNN	QNN_DLC	float	Snapdragon® X2 Elite	4.144 ms	0 - 0 MB	NPU
DnCNN	QNN_DLC	float	Snapdragon® X Elite	7.2 ms	0 - 0 MB	NPU
DnCNN	QNN_DLC	float	Snapdragon® X Elite	7.2 ms	0 - 0 MB	NPU
DnCNN	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	5.054 ms	0 - 177 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	56.0 ms	0 - 141 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	6.691 ms	0 - 2 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® SA8775P	13.889 ms	0 - 143 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® SA8775P	13.889 ms	0 - 143 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® SA8775P	13.889 ms	0 - 143 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® QCS9075	14.197 ms	0 - 2 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	13.495 ms	0 - 174 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® SA7255P	56.0 ms	0 - 141 MB	NPU
DnCNN	QNN_DLC	float	Qualcomm® SA8295P	15.306 ms	0 - 139 MB	NPU
DnCNN	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	4.029 ms	0 - 142 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.76 ms	0 - 31 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	1.221 ms	0 - 33 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® X2 Elite	1.222 ms	0 - 0 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® X Elite	2.001 ms	0 - 0 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® X Elite	2.001 ms	0 - 0 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	1.325 ms	0 - 46 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® QCS6490	7.733 ms	2 - 4 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	7.586 ms	0 - 28 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	1.794 ms	0 - 9 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® SA8775P	2.002 ms	0 - 30 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® SA8775P	2.002 ms	0 - 30 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® SA8775P	2.002 ms	0 - 30 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® QCS9075	1.933 ms	0 - 2 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® QCM6690	38.881 ms	0 - 141 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	2.384 ms	0 - 47 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® SA7255P	7.586 ms	0 - 28 MB	NPU
DnCNN	QNN_DLC	w8a8	Qualcomm® SA8295P	4.224 ms	0 - 26 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.221 ms	0 - 33 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	3.251 ms	0 - 140 MB	NPU
DnCNN	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	3.251 ms	0 - 140 MB	NPU
DnCNN	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.084 ms	0 - 148 MB	NPU
DnCNN	TFLITE	float	Snapdragon® 8 Elite Mobile	4.13 ms	0 - 143 MB	NPU
DnCNN	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	5.188 ms	0 - 179 MB	NPU
DnCNN	TFLITE	float	Qualcomm® QCS8275 (Proxy)	56.391 ms	0 - 142 MB	NPU
DnCNN	TFLITE	float	Qualcomm® QCS8550 (Proxy)	6.945 ms	0 - 2 MB	NPU
DnCNN	TFLITE	float	Qualcomm® SA8775P	14.128 ms	0 - 144 MB	NPU
DnCNN	TFLITE	float	Qualcomm® SA8775P	14.128 ms	0 - 144 MB	NPU
DnCNN	TFLITE	float	Qualcomm® SA8775P	14.128 ms	0 - 144 MB	NPU
DnCNN	TFLITE	float	Qualcomm® QCS9075	14.355 ms	0 - 4 MB	NPU
DnCNN	TFLITE	float	Qualcomm® QCS8450 (Proxy)	13.996 ms	0 - 176 MB	NPU
DnCNN	TFLITE	float	Qualcomm® SA7255P	56.391 ms	0 - 142 MB	NPU
DnCNN	TFLITE	float	Qualcomm® SA8295P	15.602 ms	0 - 141 MB	NPU
DnCNN	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	4.13 ms	0 - 143 MB	NPU
DnCNN	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.717 ms	0 - 32 MB	NPU
DnCNN	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	1.178 ms	0 - 33 MB	NPU
DnCNN	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	1.298 ms	0 - 47 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® QCS6490	7.804 ms	0 - 3 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	7.512 ms	0 - 29 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	1.725 ms	0 - 4 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® SA8775P	1.977 ms	0 - 31 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® SA8775P	1.977 ms	0 - 31 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® SA8775P	1.977 ms	0 - 31 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® QCS9075	1.85 ms	0 - 3 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® QCM6690	38.977 ms	0 - 141 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	2.338 ms	0 - 49 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® SA7255P	7.512 ms	0 - 29 MB	NPU
DnCNN	TFLITE	w8a8	Qualcomm® SA8295P	4.183 ms	0 - 27 MB	NPU
DnCNN	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.178 ms	0 - 33 MB	NPU
DnCNN	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.214 ms	0 - 145 MB	NPU
DnCNN	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.214 ms	0 - 145 MB	NPU

License

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

References

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

Paper for qualcomm/DnCNN

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

Paper • 1608.03981 • Published Aug 13, 2016