Other
PyTorch
bu_auto
android
qaihm-bot commited on
Commit
46e5eb4
·
verified ·
1 Parent(s): 5d5cf58

See https://github.com/qualcomm/ai-hub-models/releases/v0.62.0 for changelog.

Files changed (2) hide show
  1. README.md +135 -141
  2. release_assets.json +4 -4
README.md CHANGED
@@ -15,7 +15,7 @@ pipeline_tag: other
15
  BEVDet is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.
16
 
17
  This is based on the implementation of BEVDet found [here](https://github.com/HuangJunJie2017/BEVDet/).
18
- 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/bevdet) 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
 
@@ -28,23 +28,23 @@ Below are pre-exported model assets ready for deployment.
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
- | 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/bevdet/releases/v0.61.0/bevdet-onnx-float.zip)
32
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.61.0/bevdet-qnn_dlc-float.zip)
33
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.61.0/bevdet-tflite-float.zip)
34
 
35
  For more device-specific assets and performance metrics, visit **[BEVDet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/bevdet)**.
36
 
37
 
38
  ### Option 2: Export with Custom Configurations
39
 
40
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/bevdet) 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 [BEVDet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/bevdet) for usage instructions.
48
 
49
  ## Model Details
50
 
@@ -59,141 +59,135 @@ See our repository for [BEVDet on GitHub](https://github.com/qualcomm/ai-hub-mod
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
- | decoder | ONNX | float | Snapdragon® X2 Elite | 6.434 ms | 4 - 4 MB | NPU
63
- | decoder | ONNX | float | Snapdragon® X Elite | 12.463 ms | 34 - 34 MB | NPU
64
- | decoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 8.827 ms | 0 - 175 MB | NPU
65
- | decoder | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 23.574 ms | 0 - 170 MB | NPU
66
- | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 25.692 ms | 4 - 11 MB | NPU
67
- | decoder | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.954 ms | 0 - 49 MB | NPU
68
- | decoder | ONNX | float | Qualcomm® QCS8450 | 23.574 ms | 0 - 170 MB | NPU
69
- | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 23.611 ms | 4 - 11 MB | NPU
70
- | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.463 ms | 34 - 34 MB | NPU
71
- | decoder | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 7.591 ms | 3 - 151 MB | NPU
72
- | decoder | ONNX | float | Snapdragon® 8 Elite Mobile | 7.591 ms | 3 - 151 MB | NPU
73
- | decoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.567 ms | 3 - 151 MB | NPU
74
- | decoder | QNN_DLC | float | Snapdragon® X2 Elite | 6.72 ms | 4 - 4 MB | NPU
75
- | decoder | QNN_DLC | float | Snapdragon�� X Elite | 12.594 ms | 4 - 4 MB | NPU
76
- | decoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.796 ms | 0 - 170 MB | NPU
77
- | decoder | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 23.197 ms | 0 - 170 MB | NPU
78
- | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 25.793 ms | 4 - 12 MB | NPU
79
- | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 87.638 ms | 2 - 146 MB | NPU
80
- | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.909 ms | 4 - 33 MB | NPU
81
- | decoder | QNN_DLC | float | Qualcomm® SA8775P | 23.098 ms | 1 - 148 MB | NPU
82
- | decoder | QNN_DLC | float | Qualcomm® SA8650P | 23.098 ms | 1 - 148 MB | NPU
83
- | decoder | QNN_DLC | float | Qualcomm® SA8255P | 23.098 ms | 1 - 148 MB | NPU
84
- | decoder | QNN_DLC | float | Qualcomm® QCS8450 | 23.197 ms | 0 - 170 MB | NPU
85
- | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 23.712 ms | 4 - 11 MB | NPU
86
- | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.594 ms | 4 - 4 MB | NPU
87
- | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 7.608 ms | 4 - 150 MB | NPU
88
- | decoder | QNN_DLC | float | Qualcomm® SA7255P | 87.638 ms | 2 - 146 MB | NPU
89
- | decoder | QNN_DLC | float | Qualcomm® SA8295P | 25.551 ms | 0 - 142 MB | NPU
90
- | decoder | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 7.608 ms | 4 - 150 MB | NPU
91
- | decoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.581 ms | 4 - 154 MB | NPU
92
- | decoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.81 ms | 0 - 203 MB | NPU
93
- | decoder | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 23.42 ms | 1 - 205 MB | NPU
94
- | decoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 25.767 ms | 1 - 46 MB | NPU
95
- | decoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 87.575 ms | 2 - 149 MB | NPU
96
- | decoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.867 ms | 1 - 4 MB | NPU
97
- | decoder | TFLITE | float | Qualcomm® SA8775P | 23.132 ms | 1 - 152 MB | NPU
98
- | decoder | TFLITE | float | Qualcomm® SA8650P | 23.132 ms | 1 - 152 MB | NPU
99
- | decoder | TFLITE | float | Qualcomm® SA8255P | 23.132 ms | 1 - 152 MB | NPU
100
- | decoder | TFLITE | float | Qualcomm® QCS8450 | 23.42 ms | 1 - 205 MB | NPU
101
- | decoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 23.807 ms | 0 - 45 MB | NPU
102
- | decoder | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 7.551 ms | 1 - 148 MB | NPU
103
- | decoder | TFLITE | float | Qualcomm® SA7255P | 87.575 ms | 2 - 149 MB | NPU
104
- | decoder | TFLITE | float | Qualcomm® SA8295P | 25.525 ms | 1 - 146 MB | NPU
105
- | decoder | TFLITE | float | Snapdragon® 8 Elite Mobile | 7.551 ms | 1 - 148 MB | NPU
106
- | decoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.572 ms | 1 - 152 MB | NPU
107
- | encoder | ONNX | float | Snapdragon® X2 Elite | 23.552 ms | 12 - 12 MB | NPU
108
- | encoder | ONNX | float | Snapdragon® X Elite | 45.051 ms | 39 - 39 MB | NPU
109
- | encoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 36.914 ms | 19 - 442 MB | NPU
110
- | encoder | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 82.281 ms | 17 - 345 MB | NPU
111
- | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 72.585 ms | 12 - 28 MB | NPU
112
- | encoder | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 47.821 ms | 0 - 44 MB | NPU
113
- | encoder | ONNX | float | Qualcomm® QCS8450 | 82.281 ms | 17 - 345 MB | NPU
114
- | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 72.053 ms | 12 - 27 MB | NPU
115
- | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 45.051 ms | 39 - 39 MB | NPU
116
- | encoder | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 26.477 ms | 5 - 320 MB | NPU
117
- | encoder | ONNX | float | Snapdragon® 8 Elite Mobile | 26.477 ms | 5 - 320 MB | NPU
118
- | encoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.123 ms | 4 - 336 MB | NPU
119
- | encoder | QNN_DLC | float | Snapdragon® X2 Elite | 22.779 ms | 12 - 12 MB | NPU
120
- | encoder | QNN_DLC | float | Snapdragon® X Elite | 43.835 ms | 12 - 12 MB | NPU
121
- | encoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 37.42 ms | 12 - 429 MB | NPU
122
- | encoder | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 81.208 ms | 3 - 324 MB | NPU
123
- | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 67.937 ms | 12 - 29 MB | NPU
124
- | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 207.665 ms | 12 - 318 MB | NPU
125
- | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.08 ms | 13 - 18 MB | NPU
126
- | encoder | QNN_DLC | float | Qualcomm® SA8775P | 64.975 ms | 0 - 305 MB | NPU
127
- | encoder | QNN_DLC | float | Qualcomm® SA8650P | 64.975 ms | 0 - 305 MB | NPU
128
- | encoder | QNN_DLC | float | Qualcomm® SA8255P | 64.975 ms | 0 - 305 MB | NPU
129
- | encoder | QNN_DLC | float | Qualcomm® QCS8450 | 81.208 ms | 3 - 324 MB | NPU
130
- | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 65.669 ms | 14 - 30 MB | NPU
131
- | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 43.835 ms | 12 - 12 MB | NPU
132
- | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 25.26 ms | 12 - 326 MB | NPU
133
- | encoder | QNN_DLC | float | Qualcomm® SA7255P | 207.665 ms | 12 - 318 MB | NPU
134
- | encoder | QNN_DLC | float | Qualcomm® SA8295P | 68.833 ms | 0 - 231 MB | NPU
135
- | encoder | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 25.26 ms | 12 - 326 MB | NPU
136
- | encoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 20.072 ms | 11 - 347 MB | NPU
137
- | encoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 35.889 ms | 24 - 483 MB | NPU
138
- | encoder | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 75.93 ms | 25 - 390 MB | NPU
139
- | encoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 73.424 ms | 25 - 115 MB | NPU
140
- | encoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 213.601 ms | 25 - 352 MB | NPU
141
- | encoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 46.229 ms | 25 - 27 MB | NPU
142
- | encoder | TFLITE | float | Qualcomm® SA8775P | 67.296 ms | 25 - 353 MB | NPU
143
- | encoder | TFLITE | float | Qualcomm® SA8650P | 67.296 ms | 25 - 353 MB | NPU
144
- | encoder | TFLITE | float | Qualcomm® SA8255P | 67.296 ms | 25 - 353 MB | NPU
145
- | encoder | TFLITE | float | Qualcomm® QCS8450 | 75.93 ms | 25 - 390 MB | NPU
146
- | encoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 67.513 ms | 24 - 114 MB | NPU
147
- | encoder | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 28.688 ms | 25 - 359 MB | NPU
148
- | encoder | TFLITE | float | Qualcomm® SA7255P | 213.601 ms | 25 - 352 MB | NPU
149
- | encoder | TFLITE | float | Qualcomm® SA8295P | 67.93 ms | 25 - 269 MB | NPU
150
- | encoder | TFLITE | float | Snapdragon® 8 Elite Mobile | 28.688 ms | 25 - 359 MB | NPU
151
- | encoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.347 ms | 24 - 367 MB | NPU
152
- | pooler | ONNX | float | Snapdragon® X2 Elite | 21.971 ms | 535 - 535 MB | CPU
153
- | pooler | ONNX | float | Snapdragon® X Elite | 97.562 ms | 534 - 534 MB | CPU
154
- | pooler | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 26.572 ms | 332 - 339 MB | CPU
155
- | pooler | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 39.386 ms | 307 - 317 MB | CPU
156
- | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 71.144 ms | 324 - 328 MB | CPU
157
- | pooler | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 31.435 ms | 327 - 329 MB | CPU
158
- | pooler | ONNX | float | Qualcomm® QCS8450 | 39.386 ms | 307 - 317 MB | CPU
159
- | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 49.748 ms | 328 - 332 MB | CPU
160
- | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 97.562 ms | 534 - 534 MB | CPU
161
- | pooler | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 27.552 ms | 327 - 335 MB | CPU
162
- | pooler | ONNX | float | Snapdragon® 8 Elite Mobile | 27.552 ms | 327 - 335 MB | CPU
163
- | pooler | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 28.504 ms | 327 - 334 MB | CPU
164
- | pooler | QNN_DLC | float | Snapdragon® X2 Elite | 317.519 ms | 3 - 3 MB | CPU
165
- | pooler | QNN_DLC | float | Snapdragon® X Elite | 361.591 ms | 3 - 3 MB | CPU
166
- | pooler | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 154.239 ms | 409 - 425 MB | CPU
167
- | pooler | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 201.251 ms | 2 - 19 MB | CPU
168
- | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 358.204 ms | 269 - 739 MB | CPU
169
- | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 362.029 ms | 4 - 17 MB | CPU
170
- | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 171.999 ms | 7 - 11 MB | CPU
171
- | pooler | QNN_DLC | float | Qualcomm® SA8775P | 319.861 ms | 4 - 16 MB | CPU
172
- | pooler | QNN_DLC | float | Qualcomm® SA8650P | 319.861 ms | 4 - 16 MB | CPU
173
- | pooler | QNN_DLC | float | Qualcomm® SA8255P | 319.861 ms | 4 - 16 MB | CPU
174
- | pooler | QNN_DLC | float | Qualcomm® QCS8450 | 201.251 ms | 2 - 19 MB | CPU
175
- | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 333.868 ms | 270 - 740 MB | CPU
176
- | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 361.591 ms | 3 - 3 MB | CPU
177
- | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 177.937 ms | 4 - 19 MB | CPU
178
- | pooler | QNN_DLC | float | Qualcomm® SA7255P | 362.029 ms | 4 - 17 MB | CPU
179
- | pooler | QNN_DLC | float | Qualcomm® SA8295P | 311.647 ms | 3 - 15 MB | CPU
180
- | pooler | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 177.937 ms | 4 - 19 MB | CPU
181
- | pooler | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 214.032 ms | 4 - 15 MB | CPU
182
- | pooler | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 53.129 ms | 120 - 132 MB | CPU
183
- | pooler | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 81.956 ms | 128 - 143 MB | CPU
184
- | pooler | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 132.093 ms | 127 - 321 MB | CPU
185
- | pooler | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 117.49 ms | 117 - 126 MB | CPU
186
- | pooler | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 73.405 ms | 128 - 130 MB | CPU
187
- | pooler | TFLITE | float | Qualcomm® SA8775P | 101.23 ms | 128 - 133 MB | CPU
188
- | pooler | TFLITE | float | Qualcomm® SA8650P | 101.23 ms | 128 - 133 MB | CPU
189
- | pooler | TFLITE | float | Qualcomm® SA8255P | 101.23 ms | 128 - 133 MB | CPU
190
- | pooler | TFLITE | float | Qualcomm® QCS8450 | 81.956 ms | 128 - 143 MB | CPU
191
- | pooler | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 96.088 ms | 127 - 320 MB | CPU
192
- | pooler | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 56.936 ms | 127 - 139 MB | CPU
193
- | pooler | TFLITE | float | Qualcomm® SA7255P | 117.49 ms | 117 - 126 MB | CPU
194
- | pooler | TFLITE | float | Qualcomm® SA8295P | 99.538 ms | 127 - 133 MB | CPU
195
- | pooler | TFLITE | float | Snapdragon® 8 Elite Mobile | 56.936 ms | 127 - 139 MB | CPU
196
- | pooler | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 52.215 ms | 127 - 141 MB | CPU
197
 
198
  ## License
199
  * The license for the original implementation of BEVDet can be found
 
15
  BEVDet is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.
16
 
17
  This is based on the implementation of BEVDet found [here](https://github.com/HuangJunJie2017/BEVDet/).
18
+ 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/v0.62.0/src/qai_hub_models/models/bevdet) 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
 
 
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
+ | 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/bevdet/releases/v0.62.0/bevdet-onnx-float.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.62.0/bevdet-qnn_dlc-float.zip)
33
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/bevdet/releases/v0.62.0/bevdet-tflite-float.zip)
34
 
35
  For more device-specific assets and performance metrics, visit **[BEVDet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/bevdet)**.
36
 
37
 
38
  ### Option 2: Export with Custom Configurations
39
 
40
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.62.0/src/qai_hub_models/models/bevdet) 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 [BEVDet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.62.0/src/qai_hub_models/models/bevdet) for usage instructions.
48
 
49
  ## Model Details
50
 
 
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
+ | decoder | ONNX | float | Snapdragon® X2 Elite | 6.436 ms | 4 - 4 MB | NPU
63
+ | decoder | ONNX | float | Snapdragon® X Elite | 12.421 ms | 34 - 34 MB | NPU
64
+ | decoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 8.827 ms | 0 - 174 MB | NPU
65
+ | decoder | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 23.547 ms | 0 - 170 MB | NPU
66
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 25.915 ms | 4 - 11 MB | NPU
67
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.95 ms | 0 - 37 MB | NPU
68
+ | decoder | ONNX | float | Qualcomm® QCS8450 | 23.547 ms | 0 - 170 MB | NPU
69
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 23.204 ms | 4 - 11 MB | NPU
70
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.421 ms | 34 - 34 MB | NPU
71
+ | decoder | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 7.587 ms | 3 - 150 MB | NPU
72
+ | decoder | ONNX | float | Snapdragon® 8 Elite Mobile | 7.587 ms | 3 - 150 MB | NPU
73
+ | decoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.472 ms | 3 - 154 MB | NPU
74
+ | decoder | QNN_DLC | float | Snapdragon® X2 Elite | 6.757 ms | 4 - 4 MB | NPU
75
+ | decoder | QNN_DLC | float | Snapdragon® X Elite | 12.556 ms | 4 - 4 MB | NPU
76
+ | decoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.787 ms | 0 - 170 MB | NPU
77
+ | decoder | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 23.203 ms | 0 - 169 MB | NPU
78
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 25.555 ms | 4 - 12 MB | NPU
79
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.863 ms | 4 - 6 MB | NPU
80
+ | decoder | QNN_DLC | float | Qualcomm® SA8775P | 23.121 ms | 1 - 148 MB | NPU
81
+ | decoder | QNN_DLC | float | Qualcomm® SA8650P | 23.121 ms | 1 - 148 MB | NPU
82
+ | decoder | QNN_DLC | float | Qualcomm® SA8255P | 23.121 ms | 1 - 148 MB | NPU
83
+ | decoder | QNN_DLC | float | Qualcomm® QCS8450 | 23.203 ms | 0 - 169 MB | NPU
84
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 23.468 ms | 4 - 11 MB | NPU
85
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 12.556 ms | 4 - 4 MB | NPU
86
+ | decoder | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 7.56 ms | 0 - 148 MB | NPU
87
+ | decoder | QNN_DLC | float | Qualcomm® SA7255P | 87.64 ms | 1 - 145 MB | NPU
88
+ | decoder | QNN_DLC | float | Qualcomm® SA8295P | 25.535 ms | 0 - 142 MB | NPU
89
+ | decoder | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 7.56 ms | 0 - 148 MB | NPU
90
+ | decoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.567 ms | 4 - 152 MB | NPU
91
+ | decoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.828 ms | 1 - 207 MB | NPU
92
+ | decoder | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 23.206 ms | 1 - 206 MB | NPU
93
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 25.656 ms | 1 - 46 MB | NPU
94
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.971 ms | 1 - 3 MB | NPU
95
+ | decoder | TFLITE | float | Qualcomm® SA8775P | 23.126 ms | 1 - 153 MB | NPU
96
+ | decoder | TFLITE | float | Qualcomm® SA8650P | 23.126 ms | 1 - 153 MB | NPU
97
+ | decoder | TFLITE | float | Qualcomm® SA8255P | 23.126 ms | 1 - 153 MB | NPU
98
+ | decoder | TFLITE | float | Qualcomm® QCS8450 | 23.206 ms | 1 - 206 MB | NPU
99
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 23.89 ms | 1 - 46 MB | NPU
100
+ | decoder | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 7.54 ms | 0 - 148 MB | NPU
101
+ | decoder | TFLITE | float | Qualcomm® SA7255P | 87.582 ms | 1 - 149 MB | NPU
102
+ | decoder | TFLITE | float | Qualcomm® SA8295P | 25.532 ms | 1 - 146 MB | NPU
103
+ | decoder | TFLITE | float | Snapdragon® 8 Elite Mobile | 7.54 ms | 0 - 148 MB | NPU
104
+ | decoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 5.472 ms | 1 - 155 MB | NPU
105
+ | encoder | ONNX | float | Snapdragon® X2 Elite | 23.458 ms | 12 - 12 MB | NPU
106
+ | encoder | ONNX | float | Snapdragon® X Elite | 44.747 ms | 39 - 39 MB | NPU
107
+ | encoder | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 36.861 ms | 0 - 420 MB | NPU
108
+ | encoder | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 82.171 ms | 1 - 328 MB | NPU
109
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 72.746 ms | 12 - 28 MB | NPU
110
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 46.745 ms | 0 - 45 MB | NPU
111
+ | encoder | ONNX | float | Qualcomm® QCS8450 | 82.171 ms | 1 - 328 MB | NPU
112
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 67.322 ms | 12 - 27 MB | NPU
113
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 44.747 ms | 39 - 39 MB | NPU
114
+ | encoder | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 26.679 ms | 6 - 321 MB | NPU
115
+ | encoder | ONNX | float | Snapdragon® 8 Elite Mobile | 26.679 ms | 6 - 321 MB | NPU
116
+ | encoder | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.145 ms | 6 - 337 MB | NPU
117
+ | encoder | QNN_DLC | float | Snapdragon® X2 Elite | 22.822 ms | 12 - 12 MB | NPU
118
+ | encoder | QNN_DLC | float | Snapdragon® X Elite | 44.003 ms | 12 - 12 MB | NPU
119
+ | encoder | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 37.315 ms | 12 - 427 MB | NPU
120
+ | encoder | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 81.348 ms | 4 - 324 MB | NPU
121
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 67.927 ms | 12 - 29 MB | NPU
122
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.045 ms | 13 - 414 MB | NPU
123
+ | encoder | QNN_DLC | float | Qualcomm® SA8775P | 64.989 ms | 0 - 306 MB | NPU
124
+ | encoder | QNN_DLC | float | Qualcomm® SA8650P | 64.989 ms | 0 - 306 MB | NPU
125
+ | encoder | QNN_DLC | float | Qualcomm® SA8255P | 64.989 ms | 0 - 306 MB | NPU
126
+ | encoder | QNN_DLC | float | Qualcomm® QCS8450 | 81.348 ms | 4 - 324 MB | NPU
127
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 65.05 ms | 12 - 28 MB | NPU
128
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 44.003 ms | 12 - 12 MB | NPU
129
+ | encoder | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 25.259 ms | 12 - 328 MB | NPU
130
+ | encoder | QNN_DLC | float | Qualcomm® SA7255P | 207.81 ms | 1 - 306 MB | NPU
131
+ | encoder | QNN_DLC | float | Qualcomm® SA8295P | 68.789 ms | 0 - 231 MB | NPU
132
+ | encoder | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 25.259 ms | 12 - 328 MB | NPU
133
+ | encoder | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 20.119 ms | 12 - 349 MB | NPU
134
+ | encoder | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 35.917 ms | 24 - 477 MB | NPU
135
+ | encoder | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 76.238 ms | 25 - 391 MB | NPU
136
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 73.236 ms | 25 - 115 MB | NPU
137
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 45.258 ms | 25 - 27 MB | NPU
138
+ | encoder | TFLITE | float | Qualcomm® SA8775P | 67.313 ms | 25 - 353 MB | NPU
139
+ | encoder | TFLITE | float | Qualcomm® SA8650P | 67.313 ms | 25 - 353 MB | NPU
140
+ | encoder | TFLITE | float | Qualcomm® SA8255P | 67.313 ms | 25 - 353 MB | NPU
141
+ | encoder | TFLITE | float | Qualcomm® QCS8450 | 76.238 ms | 25 - 391 MB | NPU
142
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 68.202 ms | 23 - 112 MB | NPU
143
+ | encoder | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 28.858 ms | 24 - 359 MB | NPU
144
+ | encoder | TFLITE | float | Qualcomm® SA7255P | 213.598 ms | 25 - 352 MB | NPU
145
+ | encoder | TFLITE | float | Qualcomm® SA8295P | 67.796 ms | 25 - 269 MB | NPU
146
+ | encoder | TFLITE | float | Snapdragon® 8 Elite Mobile | 28.858 ms | 24 - 359 MB | NPU
147
+ | encoder | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.287 ms | 24 - 365 MB | NPU
148
+ | pooler | ONNX | float | Snapdragon® X2 Elite | 17.265 ms | 284 - 284 MB | CPU
149
+ | pooler | ONNX | float | Snapdragon® X Elite | 78.002 ms | 283 - 283 MB | CPU
150
+ | pooler | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 24.753 ms | 278 - 287 MB | CPU
151
+ | pooler | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 33.964 ms | 257 - 268 MB | CPU
152
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 54.586 ms | 274 - 278 MB | CPU
153
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 26.623 ms | 274 - 276 MB | CPU
154
+ | pooler | ONNX | float | Qualcomm® QCS8450 | 33.964 ms | 257 - 268 MB | CPU
155
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 47.101 ms | 277 - 281 MB | CPU
156
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 78.002 ms | 283 - 283 MB | CPU
157
+ | pooler | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 21.576 ms | 279 - 291 MB | CPU
158
+ | pooler | ONNX | float | Snapdragon® 8 Elite Mobile | 21.576 ms | 279 - 291 MB | CPU
159
+ | pooler | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 18.256 ms | 278 - 286 MB | CPU
160
+ | pooler | QNN_DLC | float | Snapdragon® X2 Elite | 322.292 ms | 4 - 4 MB | CPU
161
+ | pooler | QNN_DLC | float | Snapdragon® X Elite | 332.112 ms | 4 - 4 MB | CPU
162
+ | pooler | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 149.93 ms | 4 - 14 MB | CPU
163
+ | pooler | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 200.167 ms | 4 - 20 MB | CPU
164
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 369.409 ms | 268 - 736 MB | CPU
165
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 166.039 ms | 5 - 499 MB | CPU
166
+ | pooler | QNN_DLC | float | Qualcomm® SA8775P | 308.633 ms | 4 - 17 MB | CPU
167
+ | pooler | QNN_DLC | float | Qualcomm® SA8650P | 308.633 ms | 4 - 17 MB | CPU
168
+ | pooler | QNN_DLC | float | Qualcomm® SA8255P | 308.633 ms | 4 - 17 MB | CPU
169
+ | pooler | QNN_DLC | float | Qualcomm® QCS8450 | 200.167 ms | 4 - 20 MB | CPU
170
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 327.313 ms | 269 - 737 MB | CPU
171
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 332.112 ms | 4 - 4 MB | CPU
172
+ | pooler | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 173.973 ms | 4 - 17 MB | CPU
173
+ | pooler | QNN_DLC | float | Qualcomm® SA7255P | 358.012 ms | 4 - 16 MB | CPU
174
+ | pooler | QNN_DLC | float | Qualcomm® SA8295P | 306.432 ms | 4 - 16 MB | CPU
175
+ | pooler | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 173.973 ms | 4 - 17 MB | CPU
176
+ | pooler | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 215.454 ms | 4 - 13 MB | CPU
177
+ | pooler | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 46.741 ms | 127 - 140 MB | CPU
178
+ | pooler | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 74.48 ms | 120 - 134 MB | CPU
179
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 127.141 ms | 127 - 321 MB | CPU
180
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 64.365 ms | 128 - 130 MB | CPU
181
+ | pooler | TFLITE | float | Qualcomm® SA8775P | 90.951 ms | 130 - 136 MB | CPU
182
+ | pooler | TFLITE | float | Qualcomm® SA8650P | 90.951 ms | 130 - 136 MB | CPU
183
+ | pooler | TFLITE | float | Qualcomm® SA8255P | 90.951 ms | 130 - 136 MB | CPU
184
+ | pooler | TFLITE | float | Qualcomm® QCS8450 | 74.48 ms | 120 - 134 MB | CPU
185
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 106.262 ms | 127 - 320 MB | CPU
186
+ | pooler | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 52.861 ms | 126 - 138 MB | CPU
187
+ | pooler | TFLITE | float | Qualcomm® SA7255P | 108.327 ms | 128 - 137 MB | CPU
188
+ | pooler | TFLITE | float | Qualcomm® SA8295P | 86.391 ms | 127 - 133 MB | CPU
189
+ | pooler | TFLITE | float | Snapdragon® 8 Elite Mobile | 52.861 ms | 126 - 138 MB | CPU
190
+ | pooler | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 46.184 ms | 127 - 139 MB | CPU
 
 
 
 
 
 
191
 
192
  ## License
193
  * The license for the original implementation of BEVDet can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.61.0",
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/bevdet/releases/v0.61.0/bevdet-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/bevdet/releases/v0.61.0/bevdet-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/bevdet/releases/v0.61.0/bevdet-tflite-float.zip"
24
  }
25
  }
26
  }
 
1
  {
2
+ "version": "0.62.0",
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/bevdet/releases/v0.62.0/bevdet-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/bevdet/releases/v0.62.0/bevdet-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/bevdet/releases/v0.62.0/bevdet-tflite-float.zip"
24
  }
25
  }
26
  }