Hugging Face's logo

qualcomm
/
DDColor

Image-to-Image
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
0c4039f
·
verified ·
1 Parent(s): a9e91bb

v0.46.1

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.46.1 for changelog.

Files changed (5) hide show
  1. DDColor_float.dlc +0 -3
  2. DDColor_float.tflite +0 -3
  3. DDColor_w8a8.tflite +0 -3
  4. README.md +76 -236
  5. tool-versions.yaml +0 -4
DDColor_float.dlc DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4f1ba611ae0ce9e05d99a3a1b1929fc9098e7153f0e98566bbf5fb899c223f86
3
- size 225984220
 
 
 
 
DDColor_float.tflite DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a76d42c6cd76f9360f3243eb20fe1ecc12a073e754d25c1bb7a7aa395d81af3f
3
- size 225521248
 
 
 
 
DDColor_w8a8.tflite DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:dab2c126d5448782dba8f4a02dae219a8b973d7b105f149bd37ec93a3967bc64
3
- size 57456920
 
 
 
 
README.md CHANGED
@@ -9,254 +9,94 @@ pipeline_tag: image-to-image
9
 
10
  ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/web-assets/model_demo.png)
11
 
12
- # DDColor: Optimized for Mobile Deployment
13
- ## Colorize image from the black-and-white image
14
-
15
 
16
  DDColor is a coloring algorithm that produces natural, vivid color results from incoming black and white images.
17
 
18
- This model is an implementation of DDColor found [here](https://github.com/piddnad/DDColor/).
19
-
20
-
21
- This repository provides scripts to run DDColor on Qualcomm® devices.
22
- More details on model performance across various devices, can be found
23
- [here](https://aihub.qualcomm.com/models/ddcolor).
24
-
25
-
26
-
27
- ### Model Details
28
-
29
- - **Model Type:** Model_use_case.image_editing
30
- - **Model Stats:**
31
- - Model checkpoint: ddcolor_paper_tiny.pth
32
- - Input resolution: 224x224
33
- - Number of parameters: 56.3M
34
- - Model size (float): 215 MB
35
- - Model size (w8a8): 54.8 MB
36
-
37
- | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
38
- |---|---|---|---|---|---|---|---|---|
39
- | DDColor | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 249.309 ms | 1 - 1035 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
40
- | DDColor | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1959.932 ms | 1 - 568 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
41
- | DDColor | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 179.947 ms | 1 - 798 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
42
- | DDColor | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 1268.859 ms | 1 - 572 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
43
- | DDColor | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 171.384 ms | 1 - 4 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
44
- | DDColor | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 1101.236 ms | 1 - 3 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
45
- | DDColor | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 169.66 ms | 1 - 1029 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
46
- | DDColor | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1088.557 ms | 0 - 826 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
47
- | DDColor | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 249.309 ms | 1 - 1035 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
48
- | DDColor | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1959.932 ms | 1 - 568 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
49
- | DDColor | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 171.355 ms | 0 - 514 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
50
- | DDColor | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1205.131 ms | 1 - 465 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
51
- | DDColor | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 169.66 ms | 1 - 1029 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
52
- | DDColor | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1088.557 ms | 0 - 826 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
53
- | DDColor | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 117.152 ms | 1 - 1599 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
54
- | DDColor | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 825.486 ms | 1 - 1452 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
55
- | DDColor | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 83.47 ms | 1 - 745 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
56
- | DDColor | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 833.125 ms | 1 - 620 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
57
- | DDColor | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 79.012 ms | 1 - 1286 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.tflite) |
58
- | DDColor | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 683.568 ms | 3 - 752 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
59
- | DDColor | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 1139.656 ms | 1 - 1 MB | NPU | [DDColor.dlc](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor.dlc) |
60
- | DDColor | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | TFLITE | 1604.93 ms | 6 - 358 MB | CPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
61
- | DDColor | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | TFLITE | 669.754 ms | 95 - 224 MB | CPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
62
- | DDColor | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 3004.695 ms | 1 - 442 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
63
- | DDColor | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 2109.607 ms | 0 - 586 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
64
- | DDColor | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 1695.245 ms | 0 - 4 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
65
- | DDColor | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 1719.746 ms | 0 - 450 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
66
- | DDColor | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 3004.695 ms | 1 - 442 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
67
- | DDColor | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 2039.885 ms | 0 - 511 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
68
- | DDColor | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 1719.746 ms | 0 - 450 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
69
- | DDColor | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1203.501 ms | 0 - 538 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
70
- | DDColor | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 957.473 ms | 0 - 430 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
71
- | DDColor | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 476.216 ms | 41 - 372 MB | CPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
72
- | DDColor | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 706.567 ms | 2 - 507 MB | NPU | [DDColor.tflite](https://huggingface.co/qualcomm/DDColor/blob/main/DDColor_w8a8.tflite) |
73
-
74
-
75
-
76
-
77
- ## Installation
78
-
79
-
80
- Install the package via pip:
81
- ```bash
82
- # NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
83
- pip install "qai-hub-models[ddcolor]"
84
- ```
85
-
86
-
87
- ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
88
-
89
- Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
90
- Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
91
-
92
- With this API token, you can configure your client to run models on the cloud
93
- hosted devices.
94
- ```bash
95
- qai-hub configure --api_token API_TOKEN
96
- ```
97
- Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
98
-
99
-
100
-
101
- ## Demo off target
102
-
103
- The package contains a simple end-to-end demo that downloads pre-trained
104
- weights and runs this model on a sample input.
105
-
106
- ```bash
107
- python -m qai_hub_models.models.ddcolor.demo
108
- ```
109
-
110
- The above demo runs a reference implementation of pre-processing, model
111
- inference, and post processing.
112
-
113
- **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
114
- environment, please add the following to your cell (instead of the above).
115
- ```
116
- %run -m qai_hub_models.models.ddcolor.demo
117
- ```
118
-
119
-
120
- ### Run model on a cloud-hosted device
121
-
122
- In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
123
- device. This script does the following:
124
- * Performance check on-device on a cloud-hosted device
125
- * Downloads compiled assets that can be deployed on-device for Android.
126
- * Accuracy check between PyTorch and on-device outputs.
127
-
128
- ```bash
129
- python -m qai_hub_models.models.ddcolor.export
130
- ```
131
-
132
-
133
-
134
- ## How does this work?
135
-
136
- This [export script](https://aihub.qualcomm.com/models/ddcolor/qai_hub_models/models/DDColor/export.py)
137
- leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
138
- on-device. Lets go through each step below in detail:
139
-
140
- Step 1: **Compile model for on-device deployment**
141
-
142
- To compile a PyTorch model for on-device deployment, we first trace the model
143
- in memory using the `jit.trace` and then call the `submit_compile_job` API.
144
-
145
- ```python
146
- import torch
147
-
148
- import qai_hub as hub
149
- from qai_hub_models.models.ddcolor import Model
150
-
151
- # Load the model
152
- torch_model = Model.from_pretrained()
153
-
154
- # Device
155
- device = hub.Device("Samsung Galaxy S25")
156
-
157
- # Trace model
158
- input_shape = torch_model.get_input_spec()
159
- sample_inputs = torch_model.sample_inputs()
160
-
161
- pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
162
-
163
- # Compile model on a specific device
164
- compile_job = hub.submit_compile_job(
165
- model=pt_model,
166
- device=device,
167
- input_specs=torch_model.get_input_spec(),
168
- )
169
-
170
- # Get target model to run on-device
171
- target_model = compile_job.get_target_model()
172
-
173
- ```
174
-
175
-
176
- Step 2: **Performance profiling on cloud-hosted device**
177
-
178
- After compiling models from step 1. Models can be profiled model on-device using the
179
- `target_model`. Note that this scripts runs the model on a device automatically
180
- provisioned in the cloud. Once the job is submitted, you can navigate to a
181
- provided job URL to view a variety of on-device performance metrics.
182
- ```python
183
- profile_job = hub.submit_profile_job(
184
- model=target_model,
185
- device=device,
186
- )
187
-
188
- ```
189
-
190
- Step 3: **Verify on-device accuracy**
191
-
192
- To verify the accuracy of the model on-device, you can run on-device inference
193
- on sample input data on the same cloud hosted device.
194
- ```python
195
- input_data = torch_model.sample_inputs()
196
- inference_job = hub.submit_inference_job(
197
- model=target_model,
198
- device=device,
199
- inputs=input_data,
200
- )
201
- on_device_output = inference_job.download_output_data()
202
-
203
- ```
204
- With the output of the model, you can compute like PSNR, relative errors or
205
- spot check the output with expected output.
206
-
207
- **Note**: This on-device profiling and inference requires access to Qualcomm®
208
- AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
209
-
210
-
211
-
212
- ## Run demo on a cloud-hosted device
213
-
214
- You can also run the demo on-device.
215
-
216
- ```bash
217
- python -m qai_hub_models.models.ddcolor.demo --eval-mode on-device
218
- ```
219
-
220
- **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
221
- environment, please add the following to your cell (instead of the above).
222
- ```
223
- %run -m qai_hub_models.models.ddcolor.demo -- --eval-mode on-device
224
- ```
225
-
226
-
227
- ## Deploying compiled model to Android
228
-
229
-
230
- The models can be deployed using multiple runtimes:
231
- - TensorFlow Lite (`.tflite` export): [This
232
- tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
233
- guide to deploy the .tflite model in an Android application.
234
-
235
-
236
- - QNN (`.so` export ): This [sample
237
- app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
238
- provides instructions on how to use the `.so` shared library in an Android application.
239
-
240
-
241
- ## View on Qualcomm® AI Hub
242
- Get more details on DDColor's performance across various devices [here](https://aihub.qualcomm.com/models/ddcolor).
243
- Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
244
-
245
 
246
  ## License
247
  * The license for the original implementation of DDColor can be found
248
  [here](https://github.com/piddnad/DDColor/blob/master/LICENSE).
249
 
250
-
251
-
252
  ## References
253
  * [DDColor: Towards Photo-Realistic Image Colorization via Dual Decoders](https://arxiv.org/abs/2201.03545)
254
  * [Source Model Implementation](https://github.com/piddnad/DDColor/)
255
 
256
-
257
-
258
  ## Community
259
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
260
  * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
261
-
262
-
 
9
 
10
  ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/web-assets/model_demo.png)
11
 
12
+ # DDColor: Optimized for Qualcomm Devices
 
 
13
 
14
  DDColor is a coloring algorithm that produces natural, vivid color results from incoming black and white images.
15
 
16
+ This is based on the implementation of DDColor found [here](https://github.com/piddnad/DDColor/).
17
+ This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/ddcolor) 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
+
21
+ ## Getting Started
22
+ There are two ways to deploy this model on your device:
23
+
24
+ ### Option 1: Download Pre-Exported Models
25
+
26
+ Below are pre-exported model assets ready for deployment.
27
+
28
+ | Runtime | Precision | Chipset | SDK Versions | Download |
29
+ |---|---|---|---|---|
30
+ | ONNX | float | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-onnx-float.zip)
31
+ | ONNX | w8a8 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-onnx-w8a8.zip)
32
+ | ONNX | w8a8_mixed_int16 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-onnx-w8a8_mixed_int16.zip)
33
+ | QNN_DLC | float | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-qnn_dlc-float.zip)
34
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-qnn_dlc-w8a8.zip)
35
+ | QNN_DLC | w8a8_mixed_int16 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-qnn_dlc-w8a8_mixed_int16.zip)
36
+ | TFLITE | float | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-tflite-float.zip)
37
+ | TFLITE | w8a8 | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/ddcolor/releases/v0.46.1/ddcolor-tflite-w8a8.zip)
38
+
39
+ For more device-specific assets and performance metrics, visit **[DDColor on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/ddcolor)**.
40
+
41
+
42
+ ### Option 2: Export with Custom Configurations
43
+
44
+ Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/ddcolor) Python library to compile and export the model with your own:
45
+ - Custom weights (e.g., fine-tuned checkpoints)
46
+ - Custom input shapes
47
+ - Target device and runtime configurations
48
+
49
+ This option is ideal if you need to customize the model beyond the default configuration provided here.
50
+
51
+ See our repository for [DDColor on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/ddcolor) for usage instructions.
52
+
53
+ ## Model Details
54
+
55
+ **Model Type:** Model_use_case.image_editing
56
+
57
+ **Model Stats:**
58
+ - Model checkpoint: ddcolor_paper_tiny.pth
59
+ - Input resolution: 224x224
60
+ - Number of parameters: 56.3M
61
+ - Model size (float): 215 MB
62
+ - Model size (w8a8): 54.8 MB
63
+
64
+ ## Performance Summary
65
+ | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
66
+ |---|---|---|---|---|---|---
67
+ | DDColor | QNN_DLC | float | Snapdragon® X Elite | 1135.963 ms | 1 - 1 MB | NPU
68
+ | DDColor | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 826.47 ms | 0 - 1160 MB | NPU
69
+ | DDColor | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 1983.976 ms | 1 - 795 MB | NPU
70
+ | DDColor | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1100.967 ms | 1 - 4 MB | NPU
71
+ | DDColor | QNN_DLC | float | Qualcomm® SA8775P | 1101.267 ms | 1 - 644 MB | NPU
72
+ | DDColor | QNN_DLC | float | Qualcomm® QCS9075 | 1106.511 ms | 1 - 4 MB | NPU
73
+ | DDColor | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 1191.042 ms | 0 - 482 MB | NPU
74
+ | DDColor | QNN_DLC | float | Qualcomm® SA7255P | 1983.976 ms | 1 - 795 MB | NPU
75
+ | DDColor | QNN_DLC | float | Qualcomm® SA8295P | 1222.255 ms | 1 - 410 MB | NPU
76
+ | DDColor | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 832.809 ms | 1 - 754 MB | NPU
77
+ | DDColor | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 682.665 ms | 1 - 655 MB | NPU
78
+ | DDColor | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1206.113 ms | 0 - 569 MB | NPU
79
+ | DDColor | TFLITE | w8a8 | Qualcomm® QCS6490 | 678.572 ms | 106 - 235 MB | CPU
80
+ | DDColor | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 3031.978 ms | 1 - 483 MB | NPU
81
+ | DDColor | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 1697.236 ms | 0 - 4 MB | NPU
82
+ | DDColor | TFLITE | w8a8 | Qualcomm® SA8775P | 1730.938 ms | 0 - 496 MB | NPU
83
+ | DDColor | TFLITE | w8a8 | Qualcomm® QCS9075 | 1670.237 ms | 0 - 62 MB | NPU
84
+ | DDColor | TFLITE | w8a8 | Qualcomm® QCM6690 | 1694.244 ms | 9 - 409 MB | CPU
85
+ | DDColor | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 2119.822 ms | 0 - 551 MB | NPU
86
+ | DDColor | TFLITE | w8a8 | Qualcomm® SA7255P | 3031.978 ms | 1 - 483 MB | NPU
87
+ | DDColor | TFLITE | w8a8 | Qualcomm® SA8295P | 2039.833 ms | 0 - 348 MB | NPU
88
+ | DDColor | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 936.183 ms | 0 - 457 MB | NPU
89
+ | DDColor | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 487.698 ms | 4 - 374 MB | CPU
90
+ | DDColor | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 936.444 ms | 0 - 492 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
91
 
92
  ## License
93
  * The license for the original implementation of DDColor can be found
94
  [here](https://github.com/piddnad/DDColor/blob/master/LICENSE).
95
 
 
 
96
  ## References
97
  * [DDColor: Towards Photo-Realistic Image Colorization via Dual Decoders](https://arxiv.org/abs/2201.03545)
98
  * [Source Model Implementation](https://github.com/piddnad/DDColor/)
99
 
 
 
100
  ## Community
101
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
102
  * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
 
 
tool-versions.yaml DELETED
@@ -1,4 +0,0 @@
1
- tool_versions:
2
- tflite:
3
- qairt: 2.41.0.251128145156_191518
4
- tflite: 2.17.0