Hugging Face's logo

qualcomm
/
CenterPoint

Other
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
db2a5f2
·
verified ·
1 Parent(s): e424e5b

v0.60.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.60.0 for changelog.

Files changed (2) hide show
  1. README.md +44 -38
  2. release_assets.json +10 -3
README.md CHANGED
@@ -13,7 +13,7 @@ pipeline_tag: other
13
 
14
  CenterPoint is a LiDAR-based 3D object detection model that detects objects by predicting their centers and regressing other attributes. It is designed for high accuracy and real-time performance in autonomous driving applications.
15
 
16
- 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/v0.59.0/src/qai_hub_models/models/centerpoint) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
17
 
18
  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.
19
 
@@ -26,67 +26,73 @@ Below are pre-exported model assets ready for deployment.
26
 
27
  | Runtime | Precision | Chipset | SDK Versions | Download |
28
  |---|---|---|---|---|
29
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.59.0/centerpoint-qnn_dlc-float.zip)
30
- | TFLITE | float | Universal | | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.59.0/centerpoint-tflite-float.zip)
 
31
 
32
  For more device-specific assets and performance metrics, visit **[CenterPoint on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/centerpoint)**.
33
 
34
 
35
  ### Option 2: Export with Custom Configurations
36
 
37
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/centerpoint) Python library to compile and export the model with your own:
38
  - Custom weights (e.g., fine-tuned checkpoints)
39
  - Custom input shapes
40
  - Target device and runtime configurations
41
 
42
  This option is ideal if you need to customize the model beyond the default configuration provided here.
43
 
44
- See our repository for [CenterPoint on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/centerpoint) for usage instructions.
45
 
46
  ## Model Details
47
 
48
  **Model Type:** Model_use_case.driver_assistance
49
 
50
  **Model Stats:**
51
- - Model checkpoint: PointPillars
52
  - Input resolution: 5x20x5, 5x4, 5
53
- - Number of parameters: 21.8M
54
  - Model size: 83.3 MB
 
55
 
56
  ## Performance Summary
57
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
58
  |---|---|---|---|---|---|---
59
- | CenterPoint | QNN_DLC | float | Snapdragon® X2 Elite | 185.015 ms | 2 - 2 MB | NPU
60
- | CenterPoint | QNN_DLC | float | Snapdragon® X Elite | 326.336 ms | 2 - 2 MB | NPU
61
- | CenterPoint | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 249.841 ms | 0 - 750 MB | NPU
62
- | CenterPoint | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 521.899 ms | 2 - 737 MB | NPU
63
- | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 920.314 ms | 1 - 450 MB | NPU
64
- | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 332.766 ms | 2 - 6 MB | NPU
65
- | CenterPoint | QNN_DLC | float | Qualcomm® SA8775P | 397.325 ms | 1 - 703 MB | NPU
66
- | CenterPoint | QNN_DLC | float | Qualcomm® SA8650P | 397.325 ms | 1 - 703 MB | NPU
67
- | CenterPoint | QNN_DLC | float | Qualcomm® SA8255P | 397.325 ms | 1 - 703 MB | NPU
68
- | CenterPoint | QNN_DLC | float | Qualcomm® QCS8450 | 521.899 ms | 2 - 737 MB | NPU
69
- | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 388.854 ms | 4 - 13 MB | NPU
70
- | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 326.336 ms | 2 - 2 MB | NPU
71
- | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 207.78 ms | 0 - 461 MB | NPU
72
- | CenterPoint | QNN_DLC | float | Qualcomm® SA7255P | 920.314 ms | 1 - 450 MB | NPU
73
- | CenterPoint | QNN_DLC | float | Qualcomm® SA8295P | 442.028 ms | 1 - 449 MB | NPU
74
- | CenterPoint | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 207.78 ms | 0 - 461 MB | NPU
75
- | CenterPoint | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 175.341 ms | 2 - 723 MB | NPU
76
- | CenterPoint | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 4086.509 ms | 1835 - 1844 MB | CPU
77
- | CenterPoint | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5376.29 ms | 1860 - 1876 MB | CPU
78
- | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 6484.61 ms | 1847 - 1856 MB | CPU
79
- | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5056.352 ms | 1893 - 1894 MB | CPU
80
- | CenterPoint | TFLITE | float | Qualcomm® SA8775P | 5375.999 ms | 1810 - 1816 MB | CPU
81
- | CenterPoint | TFLITE | float | Qualcomm® SA8650P | 5375.999 ms | 1810 - 1816 MB | CPU
82
- | CenterPoint | TFLITE | float | Qualcomm® SA8255P | 5375.999 ms | 1810 - 1816 MB | CPU
83
- | CenterPoint | TFLITE | float | Qualcomm® QCS8450 | 5376.29 ms | 1860 - 1876 MB | CPU
84
- | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 5153.011 ms | 2363 - 2385 MB | CPU
85
- | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2647.479 ms | 1853 - 1865 MB | CPU
86
- | CenterPoint | TFLITE | float | Qualcomm® SA7255P | 6484.61 ms | 1847 - 1856 MB | CPU
87
- | CenterPoint | TFLITE | float | Qualcomm® SA8295P | 3446.214 ms | 1808 - 1814 MB | CPU
88
- | CenterPoint | TFLITE | float | Snapdragon® 8 Elite Mobile | 2647.479 ms | 1853 - 1865 MB | CPU
89
- | CenterPoint | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 803.09 ms | 1981 - 1991 MB | CPU
 
 
 
 
 
90
 
91
  ## License
92
  * The license for the original implementation of CenterPoint can be found
 
13
 
14
  CenterPoint is a LiDAR-based 3D object detection model that detects objects by predicting their centers and regressing other attributes. It is designed for high accuracy and real-time performance in autonomous driving applications.
15
 
16
+ 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/centerpoint) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
17
 
18
  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.
19
 
 
26
 
27
  | Runtime | Precision | Chipset | SDK Versions | Download |
28
  |---|---|---|---|---|
29
+ | 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/centerpoint/releases/v0.60.0/centerpoint-onnx-float.zip)
30
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.60.0/centerpoint-qnn_dlc-float.zip)
31
+ | TFLITE | float | Universal | | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.60.0/centerpoint-tflite-float.zip)
32
 
33
  For more device-specific assets and performance metrics, visit **[CenterPoint on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/centerpoint)**.
34
 
35
 
36
  ### Option 2: Export with Custom Configurations
37
 
38
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/centerpoint) Python library to compile and export the model with your own:
39
  - Custom weights (e.g., fine-tuned checkpoints)
40
  - Custom input shapes
41
  - Target device and runtime configurations
42
 
43
  This option is ideal if you need to customize the model beyond the default configuration provided here.
44
 
45
+ See our repository for [CenterPoint on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/centerpoint) for usage instructions.
46
 
47
  ## Model Details
48
 
49
  **Model Type:** Model_use_case.driver_assistance
50
 
51
  **Model Stats:**
 
52
  - Input resolution: 5x20x5, 5x4, 5
53
+ - Model checkpoint: PointPillars
54
  - Model size: 83.3 MB
55
+ - Number of parameters: 21.8M
56
 
57
  ## Performance Summary
58
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
59
  |---|---|---|---|---|---|---
60
+ | CenterPoint | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 404.732 ms | 1036 - 1057 MB | NPU
61
+ | CenterPoint | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 303.317 ms | 1056 - 1430 MB | NPU
62
+ | CenterPoint | ONNX | float | Snapdragon® 8 Elite Mobile | 303.317 ms | 1056 - 1430 MB | NPU
63
+ | CenterPoint | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 250.162 ms | 1056 - 1445 MB | NPU
64
+ | CenterPoint | QNN_DLC | float | Snapdragon® X2 Elite | 283.119 ms | 2 - 2 MB | NPU
65
+ | CenterPoint | QNN_DLC | float | Snapdragon® X Elite | 416.65 ms | 2 - 2 MB | NPU
66
+ | CenterPoint | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 307.226 ms | 3 - 692 MB | NPU
67
+ | CenterPoint | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 872.381 ms | 3 - 556 MB | NPU
68
+ | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 455.246 ms | 2 - 11 MB | NPU
69
+ | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1023.965 ms | 1 - 410 MB | NPU
70
+ | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 418.537 ms | 2 - 4 MB | NPU
71
+ | CenterPoint | QNN_DLC | float | Qualcomm® SA8775P | 480.202 ms | 2 - 408 MB | NPU
72
+ | CenterPoint | QNN_DLC | float | Qualcomm® SA8650P | 480.202 ms | 2 - 408 MB | NPU
73
+ | CenterPoint | QNN_DLC | float | Qualcomm® SA8255P | 480.202 ms | 2 - 408 MB | NPU
74
+ | CenterPoint | QNN_DLC | float | Qualcomm® QCS8450 | 872.381 ms | 3 - 556 MB | NPU
75
+ | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 471.561 ms | 4 - 13 MB | NPU
76
+ | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 416.65 ms | 2 - 2 MB | NPU
77
+ | CenterPoint | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 295.573 ms | 0 - 419 MB | NPU
78
+ | CenterPoint | QNN_DLC | float | Qualcomm® SA7255P | 1023.965 ms | 1 - 410 MB | NPU
79
+ | CenterPoint | QNN_DLC | float | Qualcomm® SA8295P | 530.256 ms | 1 - 404 MB | NPU
80
+ | CenterPoint | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 295.573 ms | 0 - 419 MB | NPU
81
+ | CenterPoint | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 253.469 ms | 1 - 411 MB | NPU
82
+ | CenterPoint | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 4073.531 ms | 1836 - 1845 MB | CPU
83
+ | CenterPoint | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5605.823 ms | 1839 - 1849 MB | CPU
84
+ | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 6099.401 ms | 2362 - 2384 MB | CPU
85
+ | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 6383.503 ms | 1839 - 1847 MB | CPU
86
+ | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5090.761 ms | 1814 - 1840 MB | CPU
87
+ | CenterPoint | TFLITE | float | Qualcomm® SA8775P | 5475.461 ms | 1810 - 1816 MB | CPU
88
+ | CenterPoint | TFLITE | float | Qualcomm® SA8650P | 5475.461 ms | 1810 - 1816 MB | CPU
89
+ | CenterPoint | TFLITE | float | Qualcomm® SA8255P | 5475.461 ms | 1810 - 1816 MB | CPU
90
+ | CenterPoint | TFLITE | float | Qualcomm® QCS8450 | 5605.823 ms | 1839 - 1849 MB | CPU
91
+ | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 5232.022 ms | 2364 - 2385 MB | CPU
92
+ | CenterPoint | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2644.894 ms | 1809 - 1818 MB | CPU
93
+ | CenterPoint | TFLITE | float | Qualcomm® SA7255P | 6383.503 ms | 1839 - 1847 MB | CPU
94
+ | CenterPoint | TFLITE | float | Snapdragon® 8 Elite Mobile | 2644.894 ms | 1809 - 1818 MB | CPU
95
+ | CenterPoint | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 856.519 ms | 1981 - 1990 MB | CPU
96
 
97
  ## License
98
  * The license for the original implementation of CenterPoint can be found
release_assets.json CHANGED
@@ -1,17 +1,24 @@
1
  {
2
- "version": "0.59.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
 
 
 
 
 
 
6
  "qnn_dlc": {
7
  "tool_versions": {
8
  "qairt": "2.45.0.260326154327"
9
  },
10
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.59.0/centerpoint-qnn_dlc-float.zip"
11
  },
12
  "tflite": {
13
  "tool_versions": {},
14
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.59.0/centerpoint-tflite-float.zip"
15
  }
16
  }
17
  }
 
1
  {
2
+ "version": "0.60.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
+ "onnx": {
7
+ "tool_versions": {
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/centerpoint/releases/v0.60.0/centerpoint-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/centerpoint/releases/v0.60.0/centerpoint-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {},
21
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/centerpoint/releases/v0.60.0/centerpoint-tflite-float.zip"
22
  }
23
  }
24
  }