Hugging Face's logo

qualcomm
/
ResNet34-SSD

Object Detection
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
37ae65e
·
verified ·
1 Parent(s): 1351d43

v0.58.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.58.0 for changelog.

Files changed (2) hide show
  1. README.md +48 -45
  2. release_assets.json +4 -4
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: object-detection
14
  ResNet34-SSD is a single-stage object detection model that integrates the ResNet34 backbone with the SSD (Single Shot MultiBox Detector) framework. It is optimized for real-time detection tasks and supports multiple deployment backends including PyTorch, TensorFlow, and ONNX.
15
 
16
  This is based on the implementation of ResNet34-SSD found [here](https://github.com/mlcommons/inference/tree/33894a19c4af6207f7cfdda75f84570f04836de5/vision/classification_and_detection).
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/v0.57.3/src/qai_hub_models/models/resnet34_ssd1200) 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
 
@@ -27,23 +27,23 @@ 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 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.57.3/resnet34_ssd1200-onnx-float.zip)
31
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.57.3/resnet34_ssd1200-qnn_dlc-float.zip)
32
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.57.3/resnet34_ssd1200-tflite-float.zip)
33
 
34
  For more device-specific assets and performance metrics, visit **[ResNet34-SSD on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/resnet34_ssd1200)**.
35
 
36
 
37
  ### Option 2: Export with Custom Configurations
38
 
39
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.57.3/src/qai_hub_models/models/resnet34_ssd1200) Python library to compile and export the model with your own:
40
  - Custom weights (e.g., fine-tuned checkpoints)
41
  - Custom input shapes
42
  - Target device and runtime configurations
43
 
44
  This option is ideal if you need to customize the model beyond the default configuration provided here.
45
 
46
- See our repository for [ResNet34-SSD on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.57.3/src/qai_hub_models/models/resnet34_ssd1200) for usage instructions.
47
 
48
  ## Model Details
49
 
@@ -58,45 +58,48 @@ See our repository for [ResNet34-SSD on GitHub](https://github.com/qualcomm/ai-h
58
  ## Performance Summary
59
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
60
  |---|---|---|---|---|---|---
61
- | ResNet34-SSD | ONNX | float | Snapdragon® X2 Elite | 43.349 ms | 164 - 164 MB | NPU
62
- | ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 88.337 ms | 132 - 132 MB | NPU
63
- | ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 63.557 ms | 17 - 507 MB | NPU
64
- | ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 177.876 ms | 17 - 437 MB | NPU
65
- | ResNet34-SSD | ONNX | float | Qualcomm® QCS8550 (Proxy) | 86.745 ms | 0 - 32 MB | NPU
66
- | ResNet34-SSD | ONNX | float | Qualcomm® QCS8450 | 177.876 ms | 17 - 437 MB | NPU
67
- | ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Mobile | 52.014 ms | 1 - 422 MB | NPU
68
- | ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 38.567 ms | 1 - 494 MB | NPU
69
- | ResNet34-SSD | ONNX | float | Qualcomm® QCS9075 | 153.507 ms | 16 - 78 MB | NPU
70
- | ResNet34-SSD | ONNX | float | Qualcomm® QCS8750 | 52.014 ms | 1 - 422 MB | NPU
71
- | ResNet34-SSD | ONNX | float | Qualcomm® QCS7181 | 88.337 ms | 132 - 132 MB | NPU
72
- | ResNet34-SSD | QNN_DLC | float | Snapdragon® X2 Elite | 62.483 ms | 17 - 17 MB | NPU
73
- | ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 129.833 ms | 17 - 17 MB | NPU
74
- | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 85.129 ms | 16 - 605 MB | NPU
75
- | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 259.163 ms | 16 - 522 MB | NPU
76
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8275 | 481.99 ms | 16 - 383 MB | NPU
77
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 134.915 ms | 17 - 19 MB | NPU
78
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8450 | 259.163 ms | 16 - 522 MB | NPU
79
- | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 67.26 ms | 16 - 391 MB | NPU
80
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8295P | 183.207 ms | 1 - 329 MB | NPU
81
- | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 51.973 ms | 0 - 547 MB | NPU
82
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA7255P | 481.99 ms | 16 - 383 MB | NPU
83
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS9075 | 194.463 ms | 17 - 35 MB | NPU
84
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8750 | 67.26 ms | 16 - 391 MB | NPU
85
- | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS7181 | 129.833 ms | 17 - 17 MB | NPU
86
- | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 107.277 ms | 0 - 542 MB | NPU
87
- | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 243.208 ms | 1 - 619 MB | NPU
88
- | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8275 | 513.425 ms | 0 - 378 MB | NPU
89
- | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 143.989 ms | 6 - 9 MB | NPU
90
- | ResNet34-SSD | TFLITE | float | Qualcomm® SA8775P | 462.59 ms | 18 - 107 MB | CPU
91
- | ResNet34-SSD | TFLITE | float | Qualcomm® SA8650P | 462.59 ms | 18 - 107 MB | CPU
92
- | ResNet34-SSD | TFLITE | float | Qualcomm® SA8255P | 462.59 ms | 18 - 107 MB | CPU
93
- | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8450 | 243.208 ms | 1 - 619 MB | NPU
94
- | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Mobile | 87.467 ms | 0 - 403 MB | NPU
95
- | ResNet34-SSD | TFLITE | float | Qualcomm® SA8295P | 201.906 ms | 0 - 353 MB | NPU
96
- | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 71.678 ms | 0 - 570 MB | NPU
97
- | ResNet34-SSD | TFLITE | float | Qualcomm® SA7255P | 513.425 ms | 0 - 378 MB | NPU
98
- | ResNet34-SSD | TFLITE | float | Qualcomm® QCS9075 | 199.528 ms | 0 - 64 MB | NPU
99
- | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8750 | 87.467 ms | 0 - 403 MB | NPU
 
 
 
100
 
101
  ## License
102
  * The license for the original implementation of ResNet34-SSD can be found
 
14
  ResNet34-SSD is a single-stage object detection model that integrates the ResNet34 backbone with the SSD (Single Shot MultiBox Detector) framework. It is optimized for real-time detection tasks and supports multiple deployment backends including PyTorch, TensorFlow, and ONNX.
15
 
16
  This is based on the implementation of ResNet34-SSD found [here](https://github.com/mlcommons/inference/tree/33894a19c4af6207f7cfdda75f84570f04836de5/vision/classification_and_detection).
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/v0.58.0/src/qai_hub_models/models/resnet34_ssd1200) 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 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.58.0/resnet34_ssd1200-onnx-float.zip)
31
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.58.0/resnet34_ssd1200-qnn_dlc-float.zip)
32
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.58.0/resnet34_ssd1200-tflite-float.zip)
33
 
34
  For more device-specific assets and performance metrics, visit **[ResNet34-SSD on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/resnet34_ssd1200)**.
35
 
36
 
37
  ### Option 2: Export with Custom Configurations
38
 
39
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/resnet34_ssd1200) Python library to compile and export the model with your own:
40
  - Custom weights (e.g., fine-tuned checkpoints)
41
  - Custom input shapes
42
  - Target device and runtime configurations
43
 
44
  This option is ideal if you need to customize the model beyond the default configuration provided here.
45
 
46
+ See our repository for [ResNet34-SSD on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/resnet34_ssd1200) for usage instructions.
47
 
48
  ## Model Details
49
 
 
58
  ## Performance Summary
59
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
60
  |---|---|---|---|---|---|---
61
+ | ResNet34-SSD | ONNX | float | Snapdragon® X2 Elite | 43.164 ms | 17 - 17 MB | NPU
62
+ | ResNet34-SSD | ONNX | float | Snapdragon® X Elite | 88.605 ms | 30 - 30 MB | NPU
63
+ | ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 63.732 ms | 0 - 502 MB | NPU
64
+ | ResNet34-SSD | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 176.341 ms | 17 - 447 MB | NPU
65
+ | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 86.854 ms | 17 - 21 MB | NPU
66
+ | ResNet34-SSD | ONNX | float | Qualcomm® QCS8450 | 176.341 ms | 17 - 447 MB | NPU
67
+ | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 135.119 ms | 16 - 36 MB | NPU
68
+ | ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 38.959 ms | 1 - 498 MB | NPU
69
+ | ResNet34-SSD | ONNX | float | Snapdragon® 8 Elite Mobile | 51.894 ms | 1 - 422 MB | NPU
70
+ | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 51.894 ms | 1 - 422 MB | NPU
71
+ | ResNet34-SSD | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 88.605 ms | 30 - 30 MB | NPU
72
+ | ResNet34-SSD | QNN_DLC | float | Snapdragon® X2 Elite | 61.754 ms | 17 - 17 MB | NPU
73
+ | ResNet34-SSD | QNN_DLC | float | Snapdragon® X Elite | 129.729 ms | 17 - 17 MB | NPU
74
+ | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 84.267 ms | 0 - 587 MB | NPU
75
+ | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 260.566 ms | 3 - 509 MB | NPU
76
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8275 | 481.817 ms | 16 - 384 MB | NPU
77
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 127.334 ms | 17 - 366 MB | NPU
78
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8775P | 173.666 ms | 7 - 376 MB | NPU
79
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8650P | 173.666 ms | 7 - 376 MB | NPU
80
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8255P | 173.666 ms | 7 - 376 MB | NPU
81
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® QCS8450 | 260.566 ms | 3 - 509 MB | NPU
82
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 172.456 ms | 17 - 35 MB | NPU
83
+ | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 52.397 ms | 14 - 559 MB | NPU
84
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA7255P | 481.817 ms | 16 - 384 MB | NPU
85
+ | ResNet34-SSD | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 67.113 ms | 9 - 385 MB | NPU
86
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® SA8295P | 183.497 ms | 0 - 329 MB | NPU
87
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 67.113 ms | 9 - 385 MB | NPU
88
+ | ResNet34-SSD | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 129.729 ms | 17 - 17 MB | NPU
89
+ | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 107.865 ms | 0 - 546 MB | NPU
90
+ | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 229.355 ms | 1 - 617 MB | NPU
91
+ | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8275 | 512.823 ms | 0 - 378 MB | NPU
92
+ | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 146.413 ms | 0 - 4 MB | NPU
93
+ | ResNet34-SSD | TFLITE | float | Qualcomm® SA8775P | 183.677 ms | 0 - 426 MB | NPU
94
+ | ResNet34-SSD | TFLITE | float | Qualcomm® SA8650P | 183.677 ms | 0 - 426 MB | NPU
95
+ | ResNet34-SSD | TFLITE | float | Qualcomm® SA8255P | 183.677 ms | 0 - 426 MB | NPU
96
+ | ResNet34-SSD | TFLITE | float | Qualcomm® QCS8450 | 229.355 ms | 1 - 617 MB | NPU
97
+ | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 181.892 ms | 0 - 65 MB | NPU
98
+ | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 72.174 ms | 0 - 570 MB | NPU
99
+ | ResNet34-SSD | TFLITE | float | Qualcomm® SA7255P | 512.823 ms | 0 - 378 MB | NPU
100
+ | ResNet34-SSD | TFLITE | float | Snapdragon® 8 Elite Mobile | 88.236 ms | 0 - 403 MB | NPU
101
+ | ResNet34-SSD | TFLITE | float | Qualcomm® SA8295P | 202.138 ms | 0 - 354 MB | NPU
102
+ | ResNet34-SSD | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 88.236 ms | 0 - 403 MB | NPU
103
 
104
  ## License
105
  * The license for the original implementation of ResNet34-SSD can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.57.3",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,20 +8,20 @@
8
  "qairt": "2.45.0.260326154327",
9
  "litert": "1.4.4"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.57.3/resnet34_ssd1200-tflite-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/resnet34_ssd1200/releases/v0.57.3/resnet34_ssd1200-qnn_dlc-float.zip"
18
  },
19
  "onnx": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327",
22
  "onnx_runtime": "1.25.0"
23
  },
24
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.57.3/resnet34_ssd1200-onnx-float.zip"
25
  }
26
  }
27
  }
 
1
  {
2
+ "version": "0.58.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
8
  "qairt": "2.45.0.260326154327",
9
  "litert": "1.4.4"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.58.0/resnet34_ssd1200-tflite-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/resnet34_ssd1200/releases/v0.58.0/resnet34_ssd1200-qnn_dlc-float.zip"
18
  },
19
  "onnx": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327",
22
  "onnx_runtime": "1.25.0"
23
  },
24
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/resnet34_ssd1200/releases/v0.58.0/resnet34_ssd1200-onnx-float.zip"
25
  }
26
  }
27
  }