Hugging Face's logo

qualcomm
/
YOLOv8-Segmentation

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
Community
1
qaihm-bot commited on
Commit
c5c89e6
·
verified ·
1 Parent(s): d799be7

v0.42.0

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.42.0 for changelog.

Files changed (1) hide show
  1. README.md +48 -46
README.md CHANGED
@@ -38,49 +38,51 @@ More details on model performance across various devices, can be found
38
 
39
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
40
  |---|---|---|---|---|---|---|---|---|
41
- | YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 16.67 ms | 4 - 85 MB | NPU | -- |
42
- | YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 16.271 ms | 0 - 127 MB | NPU | -- |
43
- | YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 8.219 ms | 4 - 50 MB | NPU | -- |
44
- | YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 12.934 ms | 5 - 42 MB | NPU | -- |
45
- | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 4.175 ms | 0 - 80 MB | NPU | -- |
46
- | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 3.986 ms | 0 - 65 MB | NPU | -- |
47
- | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 5.857 ms | 10 - 93 MB | NPU | -- |
48
- | YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.997 ms | 4 - 85 MB | NPU | -- |
49
- | YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 23.517 ms | 1 - 93 MB | NPU | -- |
50
- | YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 16.67 ms | 4 - 85 MB | NPU | -- |
51
- | YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 16.271 ms | 0 - 127 MB | NPU | -- |
52
- | YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 4.187 ms | 0 - 71 MB | NPU | -- |
53
- | YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 3.988 ms | 4 - 90 MB | NPU | -- |
54
- | YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 9.428 ms | 4 - 41 MB | NPU | -- |
55
- | YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 8.354 ms | 4 - 38 MB | NPU | -- |
56
- | YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 4.173 ms | 0 - 73 MB | NPU | -- |
57
- | YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 3.981 ms | 2 - 86 MB | NPU | -- |
58
- | YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.997 ms | 4 - 85 MB | NPU | -- |
59
- | YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 23.517 ms | 1 - 93 MB | NPU | -- |
60
- | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 3.109 ms | 0 - 155 MB | NPU | -- |
61
- | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.978 ms | 5 - 254 MB | NPU | -- |
62
- | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 3.989 ms | 0 - 109 MB | NPU | -- |
63
- | YOLOv8-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 2.46 ms | 0 - 89 MB | NPU | -- |
64
- | YOLOv8-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 2.27 ms | 5 - 108 MB | NPU | -- |
65
- | YOLOv8-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 3.33 ms | 1 - 89 MB | NPU | -- |
66
  | YOLOv8-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 1.948 ms | 0 - 86 MB | NPU | -- |
67
- | YOLOv8-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 1.807 ms | 0 - 121 MB | NPU | -- |
68
- | YOLOv8-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 2.799 ms | 3 - 76 MB | NPU | -- |
69
- | YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 4.319 ms | 129 - 129 MB | NPU | -- |
70
- | YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.064 ms | 17 - 17 MB | NPU | -- |
71
- | YOLOv8-Segmentation | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 7.321 ms | 2 - 33 MB | NPU | -- |
72
- | YOLOv8-Segmentation | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 4.72 ms | 2 - 47 MB | NPU | -- |
73
- | YOLOv8-Segmentation | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 3.704 ms | 2 - 13 MB | NPU | -- |
74
- | YOLOv8-Segmentation | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 4.319 ms | 1 - 33 MB | NPU | -- |
75
- | YOLOv8-Segmentation | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 7.321 ms | 2 - 33 MB | NPU | -- |
76
- | YOLOv8-Segmentation | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 3.704 ms | 4 - 13 MB | NPU | -- |
77
- | YOLOv8-Segmentation | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 5.032 ms | 1 - 39 MB | NPU | -- |
78
- | YOLOv8-Segmentation | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 3.688 ms | 2 - 12 MB | NPU | -- |
79
- | YOLOv8-Segmentation | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 4.319 ms | 1 - 33 MB | NPU | -- |
80
- | YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.459 ms | 2 - 42 MB | NPU | -- |
81
- | YOLOv8-Segmentation | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 1.671 ms | 2 - 41 MB | NPU | -- |
82
- | YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 1.383 ms | 2 - 42 MB | NPU | -- |
83
- | YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 4.159 ms | 5 - 5 MB | NPU | -- |
 
 
84
 
85
 
86
 
@@ -95,9 +97,9 @@ pip install "qai-hub-models[yolov8-seg]"
95
  ```
96
 
97
 
98
- ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
99
 
100
- Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
101
  Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
102
 
103
  With this API token, you can configure your client to run models on the cloud
@@ -105,7 +107,7 @@ hosted devices.
105
  ```bash
106
  qai-hub configure --api_token API_TOKEN
107
  ```
108
- Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information.
109
 
110
 
111
 
@@ -216,7 +218,7 @@ With the output of the model, you can compute like PSNR, relative errors or
216
  spot check the output with expected output.
217
 
218
  **Note**: This on-device profiling and inference requires access to Qualcomm®
219
- AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
220
 
221
 
222
 
 
38
 
39
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
40
  |---|---|---|---|---|---|---|---|---|
41
+ | YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 16.724 ms | 4 - 87 MB | NPU | -- |
42
+ | YOLOv8-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 16.226 ms | 2 - 131 MB | NPU | -- |
43
+ | YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 8.206 ms | 4 - 52 MB | NPU | -- |
44
+ | YOLOv8-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 8.183 ms | 5 - 41 MB | NPU | -- |
45
+ | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 4.16 ms | 0 - 72 MB | NPU | -- |
46
+ | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 4.03 ms | 2 - 81 MB | NPU | -- |
47
+ | YOLOv8-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 5.856 ms | 0 - 93 MB | NPU | -- |
48
+ | YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.997 ms | 4 - 87 MB | NPU | -- |
49
+ | YOLOv8-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 23.544 ms | 1 - 94 MB | NPU | -- |
50
+ | YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 16.724 ms | 4 - 87 MB | NPU | -- |
51
+ | YOLOv8-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 16.226 ms | 2 - 131 MB | NPU | -- |
52
+ | YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 4.195 ms | 0 - 83 MB | NPU | -- |
53
+ | YOLOv8-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 4.02 ms | 5 - 57 MB | NPU | -- |
54
+ | YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 9.419 ms | 4 - 42 MB | NPU | -- |
55
+ | YOLOv8-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 8.431 ms | 0 - 35 MB | NPU | -- |
56
+ | YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 4.193 ms | 0 - 71 MB | NPU | -- |
57
+ | YOLOv8-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 3.971 ms | 0 - 79 MB | NPU | -- |
58
+ | YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.997 ms | 4 - 87 MB | NPU | -- |
59
+ | YOLOv8-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 23.544 ms | 1 - 94 MB | NPU | -- |
60
+ | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 3.12 ms | 0 - 158 MB | NPU | -- |
61
+ | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.989 ms | 5 - 256 MB | NPU | -- |
62
+ | YOLOv8-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 3.976 ms | 0 - 110 MB | NPU | -- |
63
+ | YOLOv8-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 2.455 ms | 0 - 91 MB | NPU | -- |
64
+ | YOLOv8-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 2.268 ms | 5 - 107 MB | NPU | -- |
65
+ | YOLOv8-Segmentation | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 3.325 ms | 0 - 89 MB | NPU | -- |
66
  | YOLOv8-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 1.948 ms | 0 - 86 MB | NPU | -- |
67
+ | YOLOv8-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 1.81 ms | 3 - 128 MB | NPU | -- |
68
+ | YOLOv8-Segmentation | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 2.816 ms | 3 - 74 MB | NPU | -- |
69
+ | YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 4.357 ms | 130 - 130 MB | NPU | -- |
70
+ | YOLOv8-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.008 ms | 17 - 17 MB | NPU | -- |
71
+ | YOLOv8-Segmentation | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 7.45 ms | 2 - 32 MB | NPU | -- |
72
+ | YOLOv8-Segmentation | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 4.61 ms | 2 - 47 MB | NPU | -- |
73
+ | YOLOv8-Segmentation | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 3.684 ms | 2 - 13 MB | NPU | -- |
74
+ | YOLOv8-Segmentation | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 4.351 ms | 0 - 30 MB | NPU | -- |
75
+ | YOLOv8-Segmentation | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 14.935 ms | 2 - 45 MB | NPU | -- |
76
+ | YOLOv8-Segmentation | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 7.45 ms | 2 - 32 MB | NPU | -- |
77
+ | YOLOv8-Segmentation | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 3.696 ms | 2 - 13 MB | NPU | -- |
78
+ | YOLOv8-Segmentation | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 5.06 ms | 2 - 39 MB | NPU | -- |
79
+ | YOLOv8-Segmentation | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 3.705 ms | 2 - 13 MB | NPU | -- |
80
+ | YOLOv8-Segmentation | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 4.351 ms | 0 - 30 MB | NPU | -- |
81
+ | YOLOv8-Segmentation | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.462 ms | 2 - 42 MB | NPU | -- |
82
+ | YOLOv8-Segmentation | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 1.673 ms | 2 - 40 MB | NPU | -- |
83
+ | YOLOv8-Segmentation | w8a16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 4.744 ms | 2 - 41 MB | NPU | -- |
84
+ | YOLOv8-Segmentation | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 1.403 ms | 2 - 40 MB | NPU | -- |
85
+ | YOLOv8-Segmentation | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 4.101 ms | 12 - 12 MB | NPU | -- |
86
 
87
 
88
 
 
97
  ```
98
 
99
 
100
+ ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
101
 
102
+ Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
103
  Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
104
 
105
  With this API token, you can configure your client to run models on the cloud
 
107
  ```bash
108
  qai-hub configure --api_token API_TOKEN
109
  ```
110
+ Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
111
 
112
 
113
 
 
218
  spot check the output with expected output.
219
 
220
  **Note**: This on-device profiling and inference requires access to Qualcomm®
221
+ AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
222
 
223
 
224