Hugging Face's logo

qualcomm
/
YOLOv11-Segmentation

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
430badc
·
verified ·
1 Parent(s): 0863cdc

v0.48.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.48.0 for changelog.

Files changed (1) hide show
  1. README.md +30 -29
README.md CHANGED
@@ -15,18 +15,18 @@ pipeline_tag: image-segmentation
15
  Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes, segmentation masks and classes of objects in an image.
16
 
17
  This is based on the implementation of YOLOv11-Segmentation found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment).
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/yolov11_seg) 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
  Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
24
- Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) Python library to compile and export the model with your own:
25
  - Custom weights (e.g., fine-tuned checkpoints)
26
  - Custom input shapes
27
  - Target device and runtime configurations
28
 
29
- See our repository for [YOLOv11-Segmentation on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) for usage instructions.
30
 
31
 
32
  ## Model Details
@@ -44,32 +44,33 @@ See our repository for [YOLOv11-Segmentation on GitHub](https://github.com/quic/
44
  ## Performance Summary
45
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
46
  |---|---|---|---|---|---|---
47
- | YOLOv11-Segmentation | ONNX | float | Snapdragon® X Elite | 7.177 ms | 17 - 17 MB | NPU
48
- | YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.197 ms | 1 - 269 MB | NPU
49
- | YOLOv11-Segmentation | ONNX | float | Qualcomm® QCS8550 (Proxy) | 6.631 ms | 14 - 123 MB | NPU
50
- | YOLOv11-Segmentation | ONNX | float | Qualcomm® QCS9075 | 7.835 ms | 12 - 15 MB | NPU
51
- | YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.49 ms | 1 - 228 MB | NPU
52
- | YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.919 ms | 0 - 235 MB | NPU
53
- | YOLOv11-Segmentation | ONNX | float | Snapdragon® X2 Elite | 3.379 ms | 15 - 15 MB | NPU
54
- | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® X Elite | 6.43 ms | 8 - 8 MB | NPU
55
- | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.678 ms | 0 - 230 MB | NPU
56
- | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCS6490 | 425.63 ms | 165 - 170 MB | CPU
57
- | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 5.927 ms | 5 - 126 MB | NPU
58
- | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCS9075 | 7.092 ms | 6 - 9 MB | NPU
59
- | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCM6690 | 217.734 ms | 167 - 177 MB | CPU
60
- | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.743 ms | 0 - 84 MB | NPU
61
- | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 196.628 ms | 99 - 109 MB | CPU
62
- | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® X2 Elite | 2.652 ms | 6 - 6 MB | NPU
63
- | YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 3.144 ms | 0 - 117 MB | NPU
64
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 64.498 ms | 7 - 37 MB | GPU
65
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 4.293 ms | 0 - 5 MB | NPU
66
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA8775P | 6.056 ms | 4 - 90 MB | NPU
67
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS9075 | 5.882 ms | 3 - 21 MB | NPU
68
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 10.082 ms | 4 - 207 MB | NPU
69
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA7255P | 64.498 ms | 7 - 37 MB | GPU
70
- | YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA8295P | 9.375 ms | 4 - 177 MB | NPU
71
- | YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.386 ms | 0 - 93 MB | NPU
72
- | YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.937 ms | 0 - 104 MB | NPU
 
73
 
74
  ## License
75
  * The license for the original implementation of YOLOv11-Segmentation can be found
 
15
  Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes, segmentation masks and classes of objects in an image.
16
 
17
  This is based on the implementation of YOLOv11-Segmentation found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/segment).
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/main/qai_hub_models/models/yolov11_seg) 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
  Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
24
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) Python library to compile and export the model with your own:
25
  - Custom weights (e.g., fine-tuned checkpoints)
26
  - Custom input shapes
27
  - Target device and runtime configurations
28
 
29
+ See our repository for [YOLOv11-Segmentation on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/yolov11_seg) for usage instructions.
30
 
31
 
32
  ## Model Details
 
44
  ## Performance Summary
45
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
46
  |---|---|---|---|---|---|---
47
+ | YOLOv11-Segmentation | ONNX | float | Snapdragon�� X2 Elite | 3.407 ms | 16 - 16 MB | NPU
48
+ | YOLOv11-Segmentation | ONNX | float | Snapdragon® X Elite | 7.146 ms | 17 - 17 MB | NPU
49
+ | YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.196 ms | 0 - 269 MB | NPU
50
+ | YOLOv11-Segmentation | ONNX | float | Qualcomm® QCS8550 (Proxy) | 6.698 ms | 11 - 15 MB | NPU
51
+ | YOLOv11-Segmentation | ONNX | float | Qualcomm® QCS9075 | 7.837 ms | 11 - 14 MB | NPU
52
+ | YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.456 ms | 1 - 225 MB | NPU
53
+ | YOLOv11-Segmentation | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.917 ms | 0 - 235 MB | NPU
54
+ | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® X2 Elite | 2.66 ms | 6 - 6 MB | NPU
55
+ | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® X Elite | 6.411 ms | 8 - 8 MB | NPU
56
+ | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.645 ms | 8 - 236 MB | NPU
57
+ | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCS6490 | 423.428 ms | 164 - 169 MB | CPU
58
+ | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 5.887 ms | 5 - 11 MB | NPU
59
+ | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCS9075 | 7.101 ms | 7 - 10 MB | NPU
60
+ | YOLOv11-Segmentation | ONNX | w8a16 | Qualcomm® QCM6690 | 217.803 ms | 156 - 166 MB | CPU
61
+ | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.743 ms | 2 - 102 MB | NPU
62
+ | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 196.767 ms | 100 - 110 MB | CPU
63
+ | YOLOv11-Segmentation | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.425 ms | 0 - 89 MB | NPU
64
+ | YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 3.155 ms | 0 - 117 MB | NPU
65
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 15.378 ms | 4 - 85 MB | NPU
66
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 4.314 ms | 4 - 6 MB | NPU
67
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA8775P | 6.006 ms | 4 - 89 MB | NPU
68
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS9075 | 5.867 ms | 4 - 22 MB | NPU
69
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 10.077 ms | 4 - 208 MB | NPU
70
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA7255P | 15.378 ms | 4 - 85 MB | NPU
71
+ | YOLOv11-Segmentation | TFLITE | float | Qualcomm® SA8295P | 9.348 ms | 4 - 177 MB | NPU
72
+ | YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.367 ms | 0 - 95 MB | NPU
73
+ | YOLOv11-Segmentation | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.947 ms | 1 - 106 MB | NPU
74
 
75
  ## License
76
  * The license for the original implementation of YOLOv11-Segmentation can be found