ESRGAN: Optimized for Qualcomm Devices
ESRGAN is a machine learning model that upscales an image with minimal loss in quality.
This is based on the implementation of ESRGAN 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 2.51, ONNX Runtime 1.30.0 | Download |
| ONNX | w8a16 | Universal | QAIRT 2.51, ONNX Runtime 1.30.0 | Download |
| QNN_DLC | float | Universal | QAIRT 2.51 | Download |
| QNN_DLC | w8a16 | Universal | QAIRT 2.51 | Download |
| TFLITE | float | Universal | QAIRT 2.51 | Download |
For more device-specific assets and performance metrics, visit ESRGAN 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 ESRGAN on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.super_resolution
Model Stats:
- Input resolution: 128x128
- Model checkpoint: ESRGAN_x4
- Model size (float): 63.9 MB
- Number of parameters: 16.7M
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| ESRGAN | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 26.091 ms | 5 - 337 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 37.854 ms | 7 - 333 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® X2 Elite | 34.549 ms | 8 - 8 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® X Elite | 65.379 ms | 38 - 38 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 49.302 ms | 0 - 719 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 113.307 ms | 7 - 723 MB | NPU |
| ESRGAN | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 127.299 ms | 6 - 11 MB | NPU |
| ESRGAN | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 65.748 ms | 7 - 9 MB | NPU |
| ESRGAN | ONNX | float | Qualcomm® QCS8450 | 113.307 ms | 7 - 723 MB | NPU |
| ESRGAN | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 106.596 ms | 6 - 10 MB | NPU |
| ESRGAN | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 65.379 ms | 38 - 38 MB | NPU |
| ESRGAN | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 37.854 ms | 7 - 333 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 22.434 ms | 1 - 328 MB | NPU |
| ESRGAN | ONNX | float | Snapdragon® 8 Elite Gen 6 Mobile | 24.469 ms | 7 - 342 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 25.547 ms | 0 - 314 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 37.404 ms | 0 - 309 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® X2 Elite | 34.405 ms | 0 - 0 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® X Elite | 64.803 ms | 0 - 0 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 49.803 ms | 0 - 674 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 103.775 ms | 0 - 673 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 127.127 ms | 0 - 6 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 63.843 ms | 0 - 2 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® QCS8450 | 103.775 ms | 0 - 673 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 105.03 ms | 0 - 6 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 64.803 ms | 0 - 0 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 37.404 ms | 0 - 309 MB | NPU |
| ESRGAN | QNN_DLC | float | Qualcomm® SA8295P | 110.065 ms | 1 - 335 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 21.914 ms | 0 - 306 MB | NPU |
| ESRGAN | QNN_DLC | float | Snapdragon® 8 Elite Gen 6 Mobile | 23.297 ms | 0 - 317 MB | NPU |
| ESRGAN | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 25.589 ms | 0 - 345 MB | NPU |
| ESRGAN | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 38.07 ms | 3 - 348 MB | NPU |
| ESRGAN | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 48.453 ms | 39 - 756 MB | NPU |
| ESRGAN | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 116.282 ms | 4 - 710 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 127.169 ms | 3 - 47 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 65.236 ms | 3 - 5 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® SA8775P | 105.209 ms | 3 - 368 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® SA8650P | 105.209 ms | 3 - 368 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® SA8255P | 105.209 ms | 3 - 368 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® QCS8450 | 116.282 ms | 4 - 710 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 104.987 ms | 3 - 46 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 38.07 ms | 3 - 348 MB | NPU |
| ESRGAN | TFLITE | float | Qualcomm® SA7255P | 24498.945 ms | 3 - 42 MB | GPU |
| ESRGAN | TFLITE | float | Qualcomm® SA8295P | 110.102 ms | 3 - 371 MB | NPU |
| ESRGAN | TFLITE | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 21.964 ms | 3 - 346 MB | NPU |
| ESRGAN | TFLITE | float | Snapdragon® 8 Elite Gen 6 Mobile | 23.48 ms | 3 - 356 MB | NPU |
License
- The license for the original implementation of ESRGAN 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.
