NAFSSR: Stereo Image Super-Resolution Using NAFNet
Paper • 2204.08714 • Published
NAFSSR: Optimized for Qualcomm Devices
NAFNET is designed for lightweight real-time upscaling of images.
This is based on the implementation of NAFSSR 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 |
|---|---|---|---|---|
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | QAIRT 2.45 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® X2 Elite | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® X Elite | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Elite Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-8275 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-9075 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® X2 Elite | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® X Elite | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8775P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ Q-6690 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA7255P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 7 Gen 4 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Elite Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45 | Download |
For more device-specific assets and performance metrics, visit NAFSSR 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 NAFSSR on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.super_resolution
Model Stats:
- Input resolution: 128x128
- Model checkpoint: NAFSSR-S_2x
- Model size (float): 6.2 MB
- Number of parameters: 1.54M
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 79.461 ms | 8 - 8 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 199.057 ms | 41 - 41 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 127.949 ms | 4 - 11 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | 234.224 ms | 6 - 20 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 188.97 ms | 4 - 5 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8450 | 234.224 ms | 6 - 20 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 193.424 ms | 3 - 6 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 199.057 ms | 41 - 41 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 98.039 ms | 4 - 15 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | 98.039 ms | 4 - 15 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 77.756 ms | 6 - 14 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® X2 Elite | 61.091 ms | 3 - 3 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® X Elite | 183.63 ms | 33 - 33 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 113.734 ms | 2 - 9 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 147.289 ms | 2 - 5 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 174.702 ms | 0 - 34 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 176.228 ms | 2 - 4 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 183.63 ms | 33 - 33 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 1503.605 ms | 0 - 7 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 194.56 ms | 2 - 7 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 82.577 ms | 2 - 14 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 82.577 ms | 2 - 14 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 59.351 ms | 2 - 9 MB | NPU |
| NAFSSR | PRECOMPILED_QNN_ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 194.56 ms | 2 - 7 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 79.666 ms | 0 - 0 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 200.044 ms | 0 - 0 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 127.095 ms | 1 - 8 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | 235.13 ms | 1 - 14 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-8275 | 167.537 ms | 0 - 5 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-8275 | 355.369 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 188.256 ms | 1 - 2 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 195.323 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® SA8650P | 195.323 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® SA8255P | 195.323 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 | 235.13 ms | 1 - 14 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-9075 | 193.552 ms | 0 - 4 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-X7181 | 200.044 ms | 0 - 0 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ Q-8750 | 98.573 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 355.369 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 228.577 ms | 1 - 6 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | 98.573 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 77.6 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® X2 Elite | 61.373 ms | 0 - 0 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® X Elite | 183.509 ms | 0 - 0 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Gen 3 Mobile | 113.749 ms | 0 - 7 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 291.182 ms | 0 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 174.467 ms | 1 - 3 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8775P | 178.358 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8650P | 178.358 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA8255P | 178.358 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 247.919 ms | 0 - 3 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 183.509 ms | 0 - 0 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 1809.749 ms | 0 - 7 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 193.852 ms | 0 - 6 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 82.843 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Qualcomm® SA7255P | 291.182 ms | 0 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Elite Mobile | 82.843 ms | 1 - 10 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 59.329 ms | 0 - 9 MB | NPU |
| NAFSSR | QNN_CONTEXT_BINARY | w8a16 | Snapdragon® 7 Gen 4 Mobile | 193.852 ms | 0 - 6 MB | NPU |
License
- The license for the original implementation of NAFSSR 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.