Hugging Face's logo

qualcomm
/
NAFNet-DeNoise

Image-to-Image
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
7c400e4
·
verified ·
1 Parent(s): 027fe46

v0.61.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.61.0 for changelog.

Files changed (2) hide show
  1. README.md +92 -92
  2. release_assets.json +6 -6
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: pytorch
3
- license: other
4
  tags:
5
  - android
6
  pipeline_tag: image-to-image
@@ -14,7 +14,7 @@ pipeline_tag: image-to-image
14
  NAFNET is designed for lightweight real-time denoising of images.
15
 
16
  This is based on the implementation of NAFNet-DeNoise found [here](https://github.com/megvii-research/NAFNet.git).
17
- This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/nafnet_denoise) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
20
 
@@ -27,25 +27,25 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-onnx-float.zip)
31
- | ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-onnx-w8a16.zip)
32
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-qnn_dlc-float.zip)
33
- | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-qnn_dlc-w8a16.zip)
34
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-tflite-float.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[NAFNet-DeNoise on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/nafnet_denoise)**.
37
 
38
 
39
  ### Option 2: Export with Custom Configurations
40
 
41
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/nafnet_denoise) Python library to compile and export the model with your own:
42
  - Custom weights (e.g., fine-tuned checkpoints)
43
  - Custom input shapes
44
  - Target device and runtime configurations
45
 
46
  This option is ideal if you need to customize the model beyond the default configuration provided here.
47
 
48
- See our repository for [NAFNet-DeNoise on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/nafnet_denoise) for usage instructions.
49
 
50
  ## Model Details
51
 
@@ -60,89 +60,89 @@ See our repository for [NAFNet-DeNoise on GitHub](https://github.com/qualcomm/ai
60
  ## Performance Summary
61
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
62
  |---|---|---|---|---|---|---
63
- | NAFNet-DeNoise | ONNX | float | Snapdragon® X2 Elite | 14.886 ms | 4 - 4 MB | NPU
64
- | NAFNet-DeNoise | ONNX | float | Snapdragon® X Elite | 33.94 ms | 229 - 229 MB | NPU
65
- | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 27.771 ms | 0 - 846 MB | NPU
66
- | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 61.455 ms | 3 - 722 MB | NPU
67
- | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 45.536 ms | 2 - 6 MB | NPU
68
- | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 35.398 ms | 0 - 259 MB | NPU
69
- | NAFNet-DeNoise | ONNX | float | Qualcomm® QCS8450 | 61.455 ms | 3 - 722 MB | NPU
70
- | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 42.915 ms | 2 - 5 MB | NPU
71
- | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 33.94 ms | 229 - 229 MB | NPU
72
- | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 20.284 ms | 2 - 501 MB | NPU
73
- | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Elite Mobile | 20.284 ms | 2 - 501 MB | NPU
74
- | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.452 ms | 0 - 536 MB | NPU
75
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® X2 Elite | 11.773 ms | 1 - 1 MB | NPU
76
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® X Elite | 28.568 ms | 118 - 118 MB | NPU
77
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 20.573 ms | 0 - 594 MB | NPU
78
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 45.063 ms | 0 - 610 MB | NPU
79
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 132.366 ms | 1 - 4 MB | NPU
80
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 26.906 ms | 1 - 5 MB | NPU
81
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 28.425 ms | 0 - 131 MB | NPU
82
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® QCS8450 | 45.063 ms | 0 - 610 MB | NPU
83
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 27.704 ms | 1 - 4 MB | NPU
84
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 28.568 ms | 118 - 118 MB | NPU
85
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 275.972 ms | 1 - 872 MB | NPU
86
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 34.701 ms | 1 - 586 MB | NPU
87
- | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 16.362 ms | 1 - 470 MB | NPU
88
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 16.362 ms | 1 - 470 MB | NPU
89
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 11.25 ms | 1 - 548 MB | NPU
90
- | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 34.701 ms | 1 - 586 MB | NPU
91
- | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® X2 Elite | 15.742 ms | 1 - 1 MB | NPU
92
- | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® X Elite | 35.834 ms | 1 - 1 MB | NPU
93
- | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 28.346 ms | 0 - 893 MB | NPU
94
- | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 61.457 ms | 0 - 730 MB | NPU
95
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 47.34 ms | 1 - 5 MB | NPU
96
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 136.039 ms | 1 - 546 MB | NPU
97
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 37.189 ms | 1 - 3 MB | NPU
98
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8775P | 43.522 ms | 1 - 558 MB | NPU
99
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8650P | 43.522 ms | 1 - 558 MB | NPU
100
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8255P | 43.522 ms | 1 - 558 MB | NPU
101
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® QCS8450 | 61.457 ms | 0 - 730 MB | NPU
102
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 43.713 ms | 1 - 4 MB | NPU
103
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 35.834 ms | 1 - 1 MB | NPU
104
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 20.431 ms | 1 - 542 MB | NPU
105
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA7255P | 136.039 ms | 1 - 546 MB | NPU
106
- | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8295P | 48.021 ms | 1 - 398 MB | NPU
107
- | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 20.431 ms | 1 - 542 MB | NPU
108
- | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.741 ms | 0 - 622 MB | NPU
109
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 13.257 ms | 0 - 0 MB | NPU
110
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® X Elite | 28.8 ms | 1 - 1 MB | NPU
111
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 20.899 ms | 0 - 580 MB | NPU
112
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 46.019 ms | 0 - 593 MB | NPU
113
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 129.924 ms | 2 - 5 MB | NPU
114
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 26.896 ms | 0 - 3 MB | NPU
115
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 57.557 ms | 1 - 477 MB | NPU
116
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 28.453 ms | 0 - 12 MB | NPU
117
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8775P | 28.067 ms | 0 - 477 MB | NPU
118
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8650P | 28.067 ms | 0 - 477 MB | NPU
119
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8255P | 28.067 ms | 0 - 477 MB | NPU
120
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 46.019 ms | 0 - 593 MB | NPU
121
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 31.224 ms | 2 - 5 MB | NPU
122
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 28.8 ms | 1 - 1 MB | NPU
123
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 407.867 ms | 1 - 918 MB | NPU
124
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 48.236 ms | 0 - 631 MB | NPU
125
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 16.256 ms | 0 - 479 MB | NPU
126
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA7255P | 57.557 ms | 1 - 477 MB | NPU
127
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8295P | 36.897 ms | 0 - 474 MB | NPU
128
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 16.256 ms | 0 - 479 MB | NPU
129
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 11.467 ms | 0 - 520 MB | NPU
130
- | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 48.236 ms | 0 - 631 MB | NPU
131
- | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 28.396 ms | 1 - 1122 MB | NPU
132
- | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 65.385 ms | 1 - 963 MB | NPU
133
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 47.864 ms | 1 - 234 MB | NPU
134
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 136.658 ms | 1 - 765 MB | NPU
135
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 37.45 ms | 1 - 4 MB | NPU
136
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8775P | 43.744 ms | 1 - 777 MB | NPU
137
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8650P | 43.744 ms | 1 - 777 MB | NPU
138
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8255P | 43.744 ms | 1 - 777 MB | NPU
139
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® QCS8450 | 65.385 ms | 1 - 963 MB | NPU
140
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 43.706 ms | 1 - 233 MB | NPU
141
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 20.313 ms | 1 - 760 MB | NPU
142
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA7255P | 136.658 ms | 1 - 765 MB | NPU
143
- | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8295P | 50.802 ms | 1 - 629 MB | NPU
144
- | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Elite Mobile | 20.313 ms | 1 - 760 MB | NPU
145
- | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.775 ms | 1 - 850 MB | NPU
146
 
147
  ## License
148
  * The license for the original implementation of NAFNet-DeNoise can be found
 
1
  ---
2
  library_name: pytorch
3
+ license: mit
4
  tags:
5
  - android
6
  pipeline_tag: image-to-image
 
14
  NAFNET is designed for lightweight real-time denoising of images.
15
 
16
  This is based on the implementation of NAFNet-DeNoise found [here](https://github.com/megvii-research/NAFNet.git).
17
+ This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/nafnet_denoise) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
20
 
 
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
+ | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-onnx-float.zip)
31
+ | ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-onnx-w8a16.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-qnn_dlc-float.zip)
33
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-qnn_dlc-w8a16.zip)
34
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-tflite-float.zip)
35
 
36
  For more device-specific assets and performance metrics, visit **[NAFNet-DeNoise on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/nafnet_denoise)**.
37
 
38
 
39
  ### Option 2: Export with Custom Configurations
40
 
41
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/nafnet_denoise) Python library to compile and export the model with your own:
42
  - Custom weights (e.g., fine-tuned checkpoints)
43
  - Custom input shapes
44
  - Target device and runtime configurations
45
 
46
  This option is ideal if you need to customize the model beyond the default configuration provided here.
47
 
48
+ See our repository for [NAFNet-DeNoise on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/nafnet_denoise) for usage instructions.
49
 
50
  ## Model Details
51
 
 
60
  ## Performance Summary
61
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
62
  |---|---|---|---|---|---|---
63
+ | NAFNet-DeNoise | ONNX | float | Snapdragon® X2 Elite | 14.908 ms | 4 - 4 MB | NPU
64
+ | NAFNet-DeNoise | ONNX | float | Snapdragon® X Elite | 33.93 ms | 228 - 228 MB | NPU
65
+ | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 27.674 ms | 2 - 850 MB | NPU
66
+ | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 62.218 ms | 3 - 722 MB | NPU
67
+ | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 45.779 ms | 2 - 6 MB | NPU
68
+ | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 35.093 ms | 0 - 258 MB | NPU
69
+ | NAFNet-DeNoise | ONNX | float | Qualcomm® QCS8450 | 62.218 ms | 3 - 722 MB | NPU
70
+ | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 42.77 ms | 2 - 5 MB | NPU
71
+ | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 33.93 ms | 228 - 228 MB | NPU
72
+ | NAFNet-DeNoise | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 20.144 ms | 2 - 502 MB | NPU
73
+ | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Elite Mobile | 20.144 ms | 2 - 502 MB | NPU
74
+ | NAFNet-DeNoise | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.475 ms | 2 - 539 MB | NPU
75
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® X2 Elite | 11.778 ms | 1 - 1 MB | NPU
76
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® X Elite | 28.53 ms | 118 - 118 MB | NPU
77
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 20.69 ms | 0 - 594 MB | NPU
78
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 45.173 ms | 1 - 612 MB | NPU
79
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 132.744 ms | 1 - 4 MB | NPU
80
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 26.838 ms | 1 - 5 MB | NPU
81
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 28.281 ms | 0 - 131 MB | NPU
82
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® QCS8450 | 45.173 ms | 1 - 612 MB | NPU
83
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 28.049 ms | 1 - 4 MB | NPU
84
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 28.53 ms | 118 - 118 MB | NPU
85
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 266.658 ms | 0 - 870 MB | NPU
86
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 33.911 ms | 1 - 589 MB | NPU
87
+ | NAFNet-DeNoise | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 16.334 ms | 1 - 470 MB | NPU
88
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 16.334 ms | 1 - 470 MB | NPU
89
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 11.259 ms | 1 - 547 MB | NPU
90
+ | NAFNet-DeNoise | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 33.911 ms | 1 - 589 MB | NPU
91
+ | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® X2 Elite | 15.592 ms | 1 - 1 MB | NPU
92
+ | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® X Elite | 35.784 ms | 1 - 1 MB | NPU
93
+ | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 28.314 ms | 0 - 890 MB | NPU
94
+ | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 61.05 ms | 0 - 729 MB | NPU
95
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 47.424 ms | 1 - 5 MB | NPU
96
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 135.955 ms | 1 - 546 MB | NPU
97
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 38.357 ms | 1 - 16 MB | NPU
98
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8775P | 43.559 ms | 1 - 558 MB | NPU
99
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8650P | 43.559 ms | 1 - 558 MB | NPU
100
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8255P | 43.559 ms | 1 - 558 MB | NPU
101
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® QCS8450 | 61.05 ms | 0 - 729 MB | NPU
102
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 50.771 ms | 1 - 4 MB | NPU
103
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 35.784 ms | 1 - 1 MB | NPU
104
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 20.318 ms | 1 - 542 MB | NPU
105
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA7255P | 135.955 ms | 1 - 546 MB | NPU
106
+ | NAFNet-DeNoise | QNN_DLC | float | Qualcomm® SA8295P | 48.02 ms | 1 - 398 MB | NPU
107
+ | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 20.318 ms | 1 - 542 MB | NPU
108
+ | NAFNet-DeNoise | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.733 ms | 0 - 622 MB | NPU
109
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 13.258 ms | 0 - 0 MB | NPU
110
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® X Elite | 28.828 ms | 0 - 0 MB | NPU
111
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 20.829 ms | 0 - 583 MB | NPU
112
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Gen 1 Mobile | 46.392 ms | 0 - 594 MB | NPU
113
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 130.624 ms | 0 - 3 MB | NPU
114
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 26.768 ms | 0 - 3 MB | NPU
115
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 57.588 ms | 1 - 477 MB | NPU
116
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 28.637 ms | 1 - 525 MB | NPU
117
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8775P | 28.072 ms | 0 - 476 MB | NPU
118
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8650P | 28.072 ms | 0 - 476 MB | NPU
119
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8255P | 28.072 ms | 0 - 476 MB | NPU
120
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® QCS8450 | 46.392 ms | 0 - 594 MB | NPU
121
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 28.202 ms | 0 - 3 MB | NPU
122
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 28.828 ms | 0 - 0 MB | NPU
123
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 383.415 ms | 1 - 919 MB | NPU
124
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 48.499 ms | 0 - 629 MB | NPU
125
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 16.342 ms | 0 - 479 MB | NPU
126
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA7255P | 57.588 ms | 1 - 477 MB | NPU
127
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Qualcomm® SA8295P | 36.84 ms | 1 - 474 MB | NPU
128
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 16.342 ms | 0 - 479 MB | NPU
129
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 11.475 ms | 0 - 520 MB | NPU
130
+ | NAFNet-DeNoise | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 48.499 ms | 0 - 629 MB | NPU
131
+ | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 28.84 ms | 0 - 1120 MB | NPU
132
+ | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 64.184 ms | 1 - 961 MB | NPU
133
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 48.276 ms | 1 - 234 MB | NPU
134
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 136.565 ms | 1 - 764 MB | NPU
135
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 36.542 ms | 1 - 4 MB | NPU
136
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8775P | 43.698 ms | 1 - 777 MB | NPU
137
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8650P | 43.698 ms | 1 - 777 MB | NPU
138
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8255P | 43.698 ms | 1 - 777 MB | NPU
139
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® QCS8450 | 64.184 ms | 1 - 961 MB | NPU
140
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 44.022 ms | 1 - 233 MB | NPU
141
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 20.349 ms | 1 - 765 MB | NPU
142
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA7255P | 136.565 ms | 1 - 764 MB | NPU
143
+ | NAFNet-DeNoise | TFLITE | float | Qualcomm® SA8295P | 50.78 ms | 1 - 629 MB | NPU
144
+ | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Elite Mobile | 20.349 ms | 1 - 765 MB | NPU
145
+ | NAFNet-DeNoise | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.816 ms | 1 - 849 MB | NPU
146
 
147
  ## License
148
  * The license for the original implementation of NAFNet-DeNoise can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.60.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,19 +8,19 @@
8
  "qairt": "2.45.0.260326154327",
9
  "onnx_runtime": "1.27.1"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.45.0.260326154327"
16
  },
17
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327"
22
  },
23
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-tflite-float.zip"
24
  }
25
  }
26
  },
@@ -31,13 +31,13 @@
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-onnx-w8a16.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.60.0/nafnet_denoise-qnn_dlc-w8a16.zip"
41
  }
42
  }
43
  }
 
1
  {
2
+ "version": "0.61.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
8
  "qairt": "2.45.0.260326154327",
9
  "onnx_runtime": "1.27.1"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.45.0.260326154327"
16
  },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327"
22
  },
23
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-tflite-float.zip"
24
  }
25
  }
26
  },
 
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-onnx-w8a16.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/nafnet_denoise/releases/v0.61.0/nafnet_denoise-qnn_dlc-w8a16.zip"
41
  }
42
  }
43
  }