Hugging Face's logo

qualcomm
/
FastSam-S

Image Segmentation
PyTorch
android
Model card Files
xet
 Community
1
qaihm-bot commited on
Commit
3d45723
·
verified ·
1 Parent(s): fd0bc6f

v0.49.1

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.49.1 for changelog.

Files changed (1) hide show
  1. README.md +34 -34
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: image-segmentation
14
  The Fast Segment Anything Model (FastSAM) is a novel, real-time CNN-based solution for the Segment Anything task. This task is designed to segment any object within an image based on various possible user interaction prompts. The model performs competitively despite significantly reduced computation, making it a practical choice for a variety of vision tasks.
15
 
16
  This is based on the implementation of FastSam-S found [here](https://github.com/CASIA-IVA-Lab/FastSAM).
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/main/qai_hub_models/models/fastsam_s) 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 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fastsam_s/releases/v0.48.0/fastsam_s-onnx-float.zip)
31
- | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fastsam_s/releases/v0.48.0/fastsam_s-qnn_dlc-float.zip)
32
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fastsam_s/releases/v0.48.0/fastsam_s-tflite-float.zip)
33
 
34
  For more device-specific assets and performance metrics, visit **[FastSam-S on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/fastsam_s)**.
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/main/qai_hub_models/models/fastsam_s) 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 [FastSam-S on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/fastsam_s) for usage instructions.
47
 
48
  ## Model Details
49
 
@@ -59,35 +59,35 @@ See our repository for [FastSam-S on GitHub](https://github.com/qualcomm/ai-hub-
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
- | FastSam-S | ONNX | float | Snapdragon® X2 Elite | 4.363 ms | 20 - 20 MB | NPU
63
- | FastSam-S | ONNX | float | Snapdragon® X Elite | 8.549 ms | 19 - 19 MB | NPU
64
- | FastSam-S | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 5.931 ms | 14 - 281 MB | NPU
65
- | FastSam-S | ONNX | float | Qualcomm® QCS8550 (Proxy) | 8.004 ms | 0 - 27 MB | NPU
66
- | FastSam-S | ONNX | float | Qualcomm® QCS9075 | 13.117 ms | 12 - 15 MB | NPU
67
- | FastSam-S | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.494 ms | 10 - 234 MB | NPU
68
  | FastSam-S | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.367 ms | 1 - 211 MB | NPU
69
- | FastSam-S | QNN_DLC | float | Snapdragon® X2 Elite | 4.493 ms | 5 - 5 MB | NPU
70
- | FastSam-S | QNN_DLC | float | Snapdragon® X Elite | 7.947 ms | 5 - 5 MB | NPU
71
- | FastSam-S | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 5.537 ms | 0 - 209 MB | NPU
72
- | FastSam-S | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 38.657 ms | 1 - 182 MB | NPU
73
- | FastSam-S | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 7.431 ms | 5 - 7 MB | NPU
74
- | FastSam-S | QNN_DLC | float | Qualcomm® SA8775P | 11.284 ms | 1 - 188 MB | NPU
75
- | FastSam-S | QNN_DLC | float | Qualcomm® QCS9075 | 10.876 ms | 5 - 15 MB | NPU
76
- | FastSam-S | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 15.035 ms | 3 - 210 MB | NPU
77
- | FastSam-S | QNN_DLC | float | Qualcomm® SA7255P | 38.657 ms | 1 - 182 MB | NPU
78
- | FastSam-S | QNN_DLC | float | Qualcomm® SA8295P | 13.854 ms | 0 - 176 MB | NPU
79
- | FastSam-S | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.314 ms | 5 - 190 MB | NPU
80
- | FastSam-S | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.123 ms | 3 - 192 MB | NPU
81
- | FastSam-S | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.094 ms | 3 - 120 MB | NPU
82
- | FastSam-S | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 37.678 ms | 5 - 81 MB | NPU
83
- | FastSam-S | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 6.802 ms | 4 - 7 MB | NPU
84
- | FastSam-S | TFLITE | float | Qualcomm® SA8775P | 10.641 ms | 4 - 85 MB | NPU
85
- | FastSam-S | TFLITE | float | Qualcomm® QCS9075 | 10.583 ms | 4 - 39 MB | NPU
86
- | FastSam-S | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 13.983 ms | 4 - 231 MB | NPU
87
- | FastSam-S | TFLITE | float | Qualcomm® SA7255P | 37.678 ms | 5 - 81 MB | NPU
88
- | FastSam-S | TFLITE | float | Qualcomm® SA8295P | 13.113 ms | 1 - 192 MB | NPU
89
- | FastSam-S | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.835 ms | 4 - 98 MB | NPU
90
- | FastSam-S | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.889 ms | 0 - 201 MB | NPU
 
 
 
 
 
 
91
 
92
  ## License
93
  * The license for the original implementation of FastSam-S can be found
 
14
  The Fast Segment Anything Model (FastSAM) is a novel, real-time CNN-based solution for the Segment Anything task. This task is designed to segment any object within an image based on various possible user interaction prompts. The model performs competitively despite significantly reduced computation, making it a practical choice for a variety of vision tasks.
15
 
16
  This is based on the implementation of FastSam-S found [here](https://github.com/CASIA-IVA-Lab/FastSAM).
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/tree/v0.49.1/qai_hub_models/models/fastsam_s) 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 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fastsam_s/releases/v0.49.1/fastsam_s-onnx-float.zip)
31
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fastsam_s/releases/v0.49.1/fastsam_s-qnn_dlc-float.zip)
32
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fastsam_s/releases/v0.49.1/fastsam_s-tflite-float.zip)
33
 
34
  For more device-specific assets and performance metrics, visit **[FastSam-S on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/fastsam_s)**.
35
 
36
 
37
  ### Option 2: Export with Custom Configurations
38
 
39
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/fastsam_s) 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 [FastSam-S on GitHub](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/fastsam_s) for usage instructions.
47
 
48
  ## Model Details
49
 
 
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
 
 
 
 
 
 
62
  | FastSam-S | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.367 ms | 1 - 211 MB | NPU
63
+ | FastSam-S | ONNX | float | Snapdragon® X2 Elite | 4.37 ms | 20 - 20 MB | NPU
64
+ | FastSam-S | ONNX | float | Snapdragon® X Elite | 8.625 ms | 19 - 19 MB | NPU
65
+ | FastSam-S | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 5.93 ms | 15 - 285 MB | NPU
66
+ | FastSam-S | ONNX | float | Qualcomm® QCS8550 (Proxy) | 8.037 ms | 0 - 71 MB | NPU
67
+ | FastSam-S | ONNX | float | Qualcomm® QCS9075 | 13.122 ms | 11 - 14 MB | NPU
68
+ | FastSam-S | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.514 ms | 10 - 235 MB | NPU
69
+ | FastSam-S | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.114 ms | 4 - 192 MB | NPU
70
+ | FastSam-S | QNN_DLC | float | Snapdragon® X2 Elite | 4.437 ms | 5 - 5 MB | NPU
71
+ | FastSam-S | QNN_DLC | float | Snapdragon® X Elite | 7.964 ms | 5 - 5 MB | NPU
72
+ | FastSam-S | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 5.524 ms | 5 - 215 MB | NPU
73
+ | FastSam-S | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 38.665 ms | 0 - 181 MB | NPU
74
+ | FastSam-S | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 7.434 ms | 5 - 9 MB | NPU
75
+ | FastSam-S | QNN_DLC | float | Qualcomm® SA8775P | 11.279 ms | 0 - 189 MB | NPU
76
+ | FastSam-S | QNN_DLC | float | Qualcomm® QCS9075 | 10.884 ms | 5 - 15 MB | NPU
77
+ | FastSam-S | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 15.021 ms | 5 - 212 MB | NPU
78
+ | FastSam-S | QNN_DLC | float | Qualcomm® SA7255P | 38.665 ms | 0 - 181 MB | NPU
79
+ | FastSam-S | QNN_DLC | float | Qualcomm® SA8295P | 13.896 ms | 0 - 176 MB | NPU
80
+ | FastSam-S | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.297 ms | 5 - 190 MB | NPU
81
+ | FastSam-S | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.887 ms | 0 - 201 MB | NPU
82
+ | FastSam-S | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.117 ms | 3 - 122 MB | NPU
83
+ | FastSam-S | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 37.716 ms | 4 - 81 MB | NPU
84
+ | FastSam-S | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 6.817 ms | 1 - 5 MB | NPU
85
+ | FastSam-S | TFLITE | float | Qualcomm® SA8775P | 10.637 ms | 4 - 85 MB | NPU
86
+ | FastSam-S | TFLITE | float | Qualcomm® QCS9075 | 10.598 ms | 4 - 39 MB | NPU
87
+ | FastSam-S | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 13.972 ms | 2 - 230 MB | NPU
88
+ | FastSam-S | TFLITE | float | Qualcomm® SA7255P | 37.716 ms | 4 - 81 MB | NPU
89
+ | FastSam-S | TFLITE | float | Qualcomm® SA8295P | 13.104 ms | 4 - 200 MB | NPU
90
+ | FastSam-S | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.855 ms | 0 - 93 MB | NPU
91
 
92
  ## License
93
  * The license for the original implementation of FastSam-S can be found