Hugging Face's logo

qualcomm
/
DETR-ResNet101

Object Detection
PyTorch
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
adb20e5
·
verified ·
1 Parent(s): e1b1c69

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. DETR-ResNet101_float.dlc +0 -3
  2. DETR-ResNet101_float.onnx.zip +0 -3
  3. DETR-ResNet101_float.tflite +0 -3
  4. README.md +73 -227
  5. tool-versions.yaml +0 -4
DETR-ResNet101_float.dlc DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:860824d1c60cd3569c16773c04c886da881df4f0c8e94b5ccfc1235fb31a477b
3
- size 242196516
 
 
 
 
DETR-ResNet101_float.onnx.zip DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:b6c0b9ec605c08859485dbdea9890d8d0048c463c35d9c67da25063f512eb5eb
3
- size 224614212
 
 
 
 
DETR-ResNet101_float.tflite DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a3232602f0efa3b66cd16006597c55cdecb2c8c99f7ed173c012e6d4d0de393
3
- size 241546752
 
 
 
 
README.md CHANGED
@@ -10,245 +10,91 @@ pipeline_tag: object-detection
10
 
11
  ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet101/web-assets/model_demo.png)
12
 
13
- # DETR-ResNet101: Optimized for Mobile Deployment
14
- ## Transformer based object detector with ResNet101 backbone
15
-
16
 
17
  DETR is a machine learning model that can detect objects (trained on COCO dataset).
18
 
19
- This model is an implementation of DETR-ResNet101 found [here](https://github.com/facebookresearch/detr).
20
-
21
-
22
- This repository provides scripts to run DETR-ResNet101 on Qualcomm® devices.
23
- More details on model performance across various devices, can be found
24
- [here](https://aihub.qualcomm.com/models/detr_resnet101).
25
-
26
-
27
-
28
- ### Model Details
29
-
30
- - **Model Type:** Model_use_case.object_detection
31
- - **Model Stats:**
32
- - Model checkpoint: ResNet101
33
- - Input resolution: 480x480
34
- - Number of parameters: 60.3M
35
- - Model size (float): 230 MB
36
-
37
- | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
38
- |---|---|---|---|---|---|---|---|---|
39
- | DETR-ResNet101 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 137.345 ms | 0 - 339 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
40
- | DETR-ResNet101 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 133.941 ms | 5 - 292 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
41
- | DETR-ResNet101 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 54.755 ms | 0 - 379 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
42
- | DETR-ResNet101 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 55.69 ms | 5 - 337 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
43
- | DETR-ResNet101 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 26.111 ms | 0 - 4 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
44
- | DETR-ResNet101 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 24.455 ms | 5 - 7 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
45
- | DETR-ResNet101 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 24.76 ms | 0 - 127 MB | NPU | [DETR-ResNet101.onnx.zip](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.onnx.zip) |
46
- | DETR-ResNet101 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 185.671 ms | 0 - 339 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
47
- | DETR-ResNet101 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 38.807 ms | 2 - 288 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
48
- | DETR-ResNet101 | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 137.345 ms | 0 - 339 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
49
- | DETR-ResNet101 | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 133.941 ms | 5 - 292 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
50
- | DETR-ResNet101 | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 43.156 ms | 0 - 269 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
51
- | DETR-ResNet101 | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 43.513 ms | 0 - 221 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
52
- | DETR-ResNet101 | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 185.671 ms | 0 - 339 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
53
- | DETR-ResNet101 | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 38.807 ms | 2 - 288 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
54
- | DETR-ResNet101 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 18.999 ms | 0 - 475 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
55
- | DETR-ResNet101 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 17.568 ms | 5 - 486 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
56
- | DETR-ResNet101 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 18.155 ms | 4 - 395 MB | NPU | [DETR-ResNet101.onnx.zip](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.onnx.zip) |
57
- | DETR-ResNet101 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 14.546 ms | 1 - 341 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
58
- | DETR-ResNet101 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 14.508 ms | 5 - 296 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
59
- | DETR-ResNet101 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 14.531 ms | 2 - 333 MB | NPU | [DETR-ResNet101.onnx.zip](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.onnx.zip) |
60
- | DETR-ResNet101 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 11.031 ms | 0 - 371 MB | NPU | [DETR-ResNet101.tflite](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.tflite) |
61
- | DETR-ResNet101 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 11.502 ms | 5 - 309 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
62
- | DETR-ResNet101 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 11.135 ms | 5 - 276 MB | NPU | [DETR-ResNet101.onnx.zip](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.onnx.zip) |
63
- | DETR-ResNet101 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 25.372 ms | 5 - 5 MB | NPU | [DETR-ResNet101.dlc](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.dlc) |
64
- | DETR-ResNet101 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 24.786 ms | 115 - 115 MB | NPU | [DETR-ResNet101.onnx.zip](https://huggingface.co/qualcomm/DETR-ResNet101/blob/main/DETR-ResNet101.onnx.zip) |
65
-
66
-
67
-
68
-
69
- ## Installation
70
-
71
-
72
- Install the package via pip:
73
- ```bash
74
- # NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
75
- pip install "qai-hub-models[detr-resnet101]"
76
- ```
77
-
78
-
79
- ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
80
-
81
- Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
82
- Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
83
-
84
- With this API token, you can configure your client to run models on the cloud
85
- hosted devices.
86
- ```bash
87
- qai-hub configure --api_token API_TOKEN
88
- ```
89
- Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
90
-
91
-
92
-
93
- ## Demo off target
94
-
95
- The package contains a simple end-to-end demo that downloads pre-trained
96
- weights and runs this model on a sample input.
97
-
98
- ```bash
99
- python -m qai_hub_models.models.detr_resnet101.demo
100
- ```
101
-
102
- The above demo runs a reference implementation of pre-processing, model
103
- inference, and post processing.
104
-
105
- **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
106
- environment, please add the following to your cell (instead of the above).
107
- ```
108
- %run -m qai_hub_models.models.detr_resnet101.demo
109
- ```
110
-
111
-
112
- ### Run model on a cloud-hosted device
113
-
114
- In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
115
- device. This script does the following:
116
- * Performance check on-device on a cloud-hosted device
117
- * Downloads compiled assets that can be deployed on-device for Android.
118
- * Accuracy check between PyTorch and on-device outputs.
119
-
120
- ```bash
121
- python -m qai_hub_models.models.detr_resnet101.export
122
- ```
123
-
124
-
125
-
126
- ## How does this work?
127
-
128
- This [export script](https://aihub.qualcomm.com/models/detr_resnet101/qai_hub_models/models/DETR-ResNet101/export.py)
129
- leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
130
- on-device. Lets go through each step below in detail:
131
-
132
- Step 1: **Compile model for on-device deployment**
133
-
134
- To compile a PyTorch model for on-device deployment, we first trace the model
135
- in memory using the `jit.trace` and then call the `submit_compile_job` API.
136
-
137
- ```python
138
- import torch
139
-
140
- import qai_hub as hub
141
- from qai_hub_models.models.detr_resnet101 import Model
142
-
143
- # Load the model
144
- torch_model = Model.from_pretrained()
145
-
146
- # Device
147
- device = hub.Device("Samsung Galaxy S25")
148
-
149
- # Trace model
150
- input_shape = torch_model.get_input_spec()
151
- sample_inputs = torch_model.sample_inputs()
152
-
153
- pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
154
-
155
- # Compile model on a specific device
156
- compile_job = hub.submit_compile_job(
157
- model=pt_model,
158
- device=device,
159
- input_specs=torch_model.get_input_spec(),
160
- )
161
-
162
- # Get target model to run on-device
163
- target_model = compile_job.get_target_model()
164
-
165
- ```
166
-
167
-
168
- Step 2: **Performance profiling on cloud-hosted device**
169
-
170
- After compiling models from step 1. Models can be profiled model on-device using the
171
- `target_model`. Note that this scripts runs the model on a device automatically
172
- provisioned in the cloud. Once the job is submitted, you can navigate to a
173
- provided job URL to view a variety of on-device performance metrics.
174
- ```python
175
- profile_job = hub.submit_profile_job(
176
- model=target_model,
177
- device=device,
178
- )
179
-
180
- ```
181
-
182
- Step 3: **Verify on-device accuracy**
183
-
184
- To verify the accuracy of the model on-device, you can run on-device inference
185
- on sample input data on the same cloud hosted device.
186
- ```python
187
- input_data = torch_model.sample_inputs()
188
- inference_job = hub.submit_inference_job(
189
- model=target_model,
190
- device=device,
191
- inputs=input_data,
192
- )
193
- on_device_output = inference_job.download_output_data()
194
-
195
- ```
196
- With the output of the model, you can compute like PSNR, relative errors or
197
- spot check the output with expected output.
198
-
199
- **Note**: This on-device profiling and inference requires access to Qualcomm®
200
- AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
201
-
202
-
203
-
204
- ## Run demo on a cloud-hosted device
205
-
206
- You can also run the demo on-device.
207
-
208
- ```bash
209
- python -m qai_hub_models.models.detr_resnet101.demo --eval-mode on-device
210
- ```
211
-
212
- **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
213
- environment, please add the following to your cell (instead of the above).
214
- ```
215
- %run -m qai_hub_models.models.detr_resnet101.demo -- --eval-mode on-device
216
- ```
217
-
218
-
219
- ## Deploying compiled model to Android
220
-
221
-
222
- The models can be deployed using multiple runtimes:
223
- - TensorFlow Lite (`.tflite` export): [This
224
- tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
225
- guide to deploy the .tflite model in an Android application.
226
-
227
-
228
- - QNN (`.so` export ): This [sample
229
- app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
230
- provides instructions on how to use the `.so` shared library in an Android application.
231
-
232
-
233
- ## View on Qualcomm® AI Hub
234
- Get more details on DETR-ResNet101's performance across various devices [here](https://aihub.qualcomm.com/models/detr_resnet101).
235
- Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
236
-
237
 
238
  ## License
239
  * The license for the original implementation of DETR-ResNet101 can be found
240
  [here](https://github.com/facebookresearch/detr/blob/main/LICENSE).
241
 
242
-
243
-
244
  ## References
245
  * [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872)
246
  * [Source Model Implementation](https://github.com/facebookresearch/detr)
247
 
248
-
249
-
250
  ## Community
251
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
252
  * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
253
-
254
-
 
10
 
11
  ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet101/web-assets/model_demo.png)
12
 
13
+ # DETR-ResNet101: Optimized for Qualcomm Devices
 
 
14
 
15
  DETR is a machine learning model that can detect objects (trained on COCO dataset).
16
 
17
+ This is based on the implementation of DETR-ResNet101 found [here](https://github.com/facebookresearch/detr).
18
+ 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/detr_resnet101) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
19
+
20
+ 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.
21
+
22
+ ## Getting Started
23
+ There are two ways to deploy this model on your device:
24
+
25
+ ### Option 1: Download Pre-Exported Models
26
+
27
+ Below are pre-exported model assets ready for deployment.
28
+
29
+ | Runtime | Precision | Chipset | SDK Versions | Download |
30
+ |---|---|---|---|---|
31
+ | 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/detr_resnet101/releases/v0.46.1/detr_resnet101-onnx-float.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/detr_resnet101/releases/v0.46.1/detr_resnet101-qnn_dlc-float.zip)
33
+ | 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/detr_resnet101/releases/v0.46.1/detr_resnet101-tflite-float.zip)
34
+
35
+ For more device-specific assets and performance metrics, visit **[DETR-ResNet101 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/detr_resnet101)**.
36
+
37
+
38
+ ### Option 2: Export with Custom Configurations
39
+
40
+ Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/detr_resnet101) Python library to compile and export the model with your own:
41
+ - Custom weights (e.g., fine-tuned checkpoints)
42
+ - Custom input shapes
43
+ - Target device and runtime configurations
44
+
45
+ This option is ideal if you need to customize the model beyond the default configuration provided here.
46
+
47
+ See our repository for [DETR-ResNet101 on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/detr_resnet101) for usage instructions.
48
+
49
+ ## Model Details
50
+
51
+ **Model Type:** Model_use_case.object_detection
52
+
53
+ **Model Stats:**
54
+ - Model checkpoint: ResNet101
55
+ - Input resolution: 480x480
56
+ - Number of parameters: 60.3M
57
+ - Model size (float): 230 MB
58
+
59
+ ## Performance Summary
60
+ | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
+ |---|---|---|---|---|---|---
62
+ | DETR-ResNet101 | ONNX | float | Snapdragon® X Elite | 24.803 ms | 115 - 115 MB | NPU
63
+ | DETR-ResNet101 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 18.062 ms | 1 - 451 MB | NPU
64
+ | DETR-ResNet101 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 24.774 ms | 0 - 3 MB | NPU
65
+ | DETR-ResNet101 | ONNX | float | Qualcomm® QCS9075 | 41.363 ms | 5 - 12 MB | NPU
66
+ | DETR-ResNet101 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 14.941 ms | 1 - 332 MB | NPU
67
+ | DETR-ResNet101 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.12 ms | 5 - 278 MB | NPU
68
+ | DETR-ResNet101 | QNN_DLC | float | Snapdragon® X Elite | 27.117 ms | 5 - 5 MB | NPU
69
+ | DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 19.81 ms | 5 - 471 MB | NPU
70
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 138.374 ms | 1 - 335 MB | NPU
71
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 26.71 ms | 5 - 7 MB | NPU
72
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA8775P | 41.241 ms | 0 - 334 MB | NPU
73
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® QCS9075 | 46.187 ms | 5 - 11 MB | NPU
74
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 56.534 ms | 2 - 364 MB | NPU
75
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA7255P | 138.374 ms | 1 - 335 MB | NPU
76
+ | DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA8295P | 43.987 ms | 0 - 251 MB | NPU
77
+ | DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 14.805 ms | 1 - 342 MB | NPU
78
+ | DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.74 ms | 5 - 361 MB | NPU
79
+ | DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 19.793 ms | 0 - 537 MB | NPU
80
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 138.24 ms | 0 - 403 MB | NPU
81
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 26.657 ms | 0 - 3 MB | NPU
82
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® SA8775P | 41.124 ms | 0 - 400 MB | NPU
83
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® QCS9075 | 41.507 ms | 0 - 125 MB | NPU
84
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 54.877 ms | 0 - 420 MB | NPU
85
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® SA7255P | 138.24 ms | 0 - 403 MB | NPU
86
+ | DETR-ResNet101 | TFLITE | float | Qualcomm® SA8295P | 43.769 ms | 0 - 307 MB | NPU
87
+ | DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 15.264 ms | 0 - 411 MB | NPU
88
+ | DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.203 ms | 0 - 426 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
89
 
90
  ## License
91
  * The license for the original implementation of DETR-ResNet101 can be found
92
  [here](https://github.com/facebookresearch/detr/blob/main/LICENSE).
93
 
 
 
94
  ## References
95
  * [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872)
96
  * [Source Model Implementation](https://github.com/facebookresearch/detr)
97
 
 
 
98
  ## Community
99
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
100
  * 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
- onnx:
3
- qairt: 2.37.1.250807093845_124904
4
- onnx_runtime: 1.23.0