Hugging Face's logo

qualcomm
/
SINet

Image Segmentation
PyTorch
TF Lite
ONNX
android
Model card Files Community
qaihm-bot commited on
Commit
71c1d82
·
verified ·
1 Parent(s): a177dd5

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +33 -32
README.md CHANGED
@@ -35,39 +35,39 @@ More details on model performance across various devices, can be found
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.764 ms | 0 - 6 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
39
- | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.18 ms | 1 - 7 MB | FP16 | NPU | [SINet.so](https://huggingface.co/qualcomm/SINet/blob/main/SINet.so) |
40
- | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.292 ms | 0 - 8 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
41
- | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.152 ms | 0 - 23 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
42
- | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.798 ms | 0 - 26 MB | FP16 | NPU | [SINet.so](https://huggingface.co/qualcomm/SINet/blob/main/SINet.so) |
43
- | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.569 ms | 0 - 24 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
44
- | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.957 ms | 0 - 20 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
45
- | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.747 ms | 0 - 20 MB | FP16 | NPU | Use Export Script |
46
- | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.559 ms | 0 - 20 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
47
- | SINet | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.76 ms | 0 - 7 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
48
- | SINet | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 1.166 ms | 1 - 5 MB | FP16 | NPU | Use Export Script |
49
- | SINet | SA7255P ADP | SA7255P | TFLITE | 9.057 ms | 0 - 13 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
50
- | SINet | SA7255P ADP | SA7255P | QNN | 8.059 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
51
- | SINet | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.76 ms | 0 - 6 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
52
- | SINet | SA8255 (Proxy) | SA8255P Proxy | QNN | 1.182 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
53
- | SINet | SA8295P ADP | SA8295P | TFLITE | 2.411 ms | 0 - 19 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
54
- | SINet | SA8295P ADP | SA8295P | QNN | 2.075 ms | 0 - 14 MB | FP16 | NPU | Use Export Script |
55
- | SINet | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.756 ms | 0 - 6 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
56
- | SINet | SA8650 (Proxy) | SA8650P Proxy | QNN | 1.181 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
57
- | SINet | SA8775P ADP | SA8775P | TFLITE | 2.697 ms | 0 - 13 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
58
- | SINet | SA8775P ADP | SA8775P | QNN | 1.987 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
59
- | SINet | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.889 ms | 0 - 21 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
60
- | SINet | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.322 ms | 1 - 19 MB | FP16 | NPU | Use Export Script |
61
- | SINet | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.358 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
62
- | SINet | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.393 ms | 2 - 2 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
69
- This model can be installed as a Python package via pip.
70
 
 
71
  ```bash
72
  pip install qai-hub-models
73
  ```
@@ -123,10 +123,10 @@ Profiling Results
123
  SINet
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
- Estimated inference time (ms) : 1.8
127
- Estimated peak memory usage (MB): [0, 6]
128
- Total # Ops : 240
129
- Compute Unit(s) : NPU (240 ops)
130
  ```
131
 
132
 
@@ -151,7 +151,7 @@ from qai_hub_models.models.sinet import Model
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
- device = hub.Device("Samsung Galaxy S23")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
@@ -243,7 +243,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
243
 
244
 
245
  ## License
246
- * The license for the original implementation of SINet can be found [here](https://github.com/clovaai/ext_portrait_segmentation/blob/master/LICENSE).
 
247
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
248
 
249
 
 
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.223 ms | 0 - 5 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
39
+ | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.656 ms | 1 - 6 MB | FP16 | NPU | [SINet.so](https://huggingface.co/qualcomm/SINet/blob/main/SINet.so) |
40
+ | SINet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.396 ms | 0 - 8 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
41
+ | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.449 ms | 0 - 24 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
42
+ | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.103 ms | 0 - 29 MB | FP16 | NPU | [SINet.so](https://huggingface.co/qualcomm/SINet/blob/main/SINet.so) |
43
+ | SINet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.569 ms | 1 - 30 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
44
+ | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.192 ms | 0 - 20 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
45
+ | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.046 ms | 1 - 21 MB | FP16 | NPU | Use Export Script |
46
+ | SINet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.325 ms | 1 - 21 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
47
+ | SINet | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.222 ms | 0 - 8 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
48
+ | SINet | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 1.616 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
49
+ | SINet | SA7255P ADP | SA7255P | TFLITE | 10.736 ms | 0 - 14 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
50
+ | SINet | SA7255P ADP | SA7255P | QNN | 9.697 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
51
+ | SINet | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.227 ms | 0 - 6 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
52
+ | SINet | SA8255 (Proxy) | SA8255P Proxy | QNN | 1.614 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
53
+ | SINet | SA8295P ADP | SA8295P | TFLITE | 3.078 ms | 0 - 18 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
54
+ | SINet | SA8295P ADP | SA8295P | QNN | 2.627 ms | 0 - 15 MB | FP16 | NPU | Use Export Script |
55
+ | SINet | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.224 ms | 0 - 7 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
56
+ | SINet | SA8650 (Proxy) | SA8650P Proxy | QNN | 1.614 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
57
+ | SINet | SA8775P ADP | SA8775P | TFLITE | 3.277 ms | 0 - 13 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
58
+ | SINet | SA8775P ADP | SA8775P | QNN | 2.608 ms | 1 - 11 MB | FP16 | NPU | Use Export Script |
59
+ | SINet | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 2.418 ms | 0 - 26 MB | FP16 | NPU | [SINet.tflite](https://huggingface.co/qualcomm/SINet/blob/main/SINet.tflite) |
60
+ | SINet | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 1.917 ms | 1 - 27 MB | FP16 | NPU | Use Export Script |
61
+ | SINet | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.836 ms | 1 - 1 MB | FP16 | NPU | Use Export Script |
62
+ | SINet | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.606 ms | 2 - 2 MB | FP16 | NPU | [SINet.onnx](https://huggingface.co/qualcomm/SINet/blob/main/SINet.onnx) |
63
 
64
 
65
 
66
 
67
  ## Installation
68
 
 
69
 
70
+ Install the package via pip:
71
  ```bash
72
  pip install qai-hub-models
73
  ```
 
123
  SINet
124
  Device : Samsung Galaxy S23 (13)
125
  Runtime : TFLITE
126
+ Estimated inference time (ms) : 2.2
127
+ Estimated peak memory usage (MB): [0, 5]
128
+ Total # Ops : 242
129
+ Compute Unit(s) : NPU (242 ops)
130
  ```
131
 
132
 
 
151
  torch_model = Model.from_pretrained()
152
 
153
  # Device
154
+ device = hub.Device("Samsung Galaxy S24")
155
 
156
  # Trace model
157
  input_shape = torch_model.get_input_spec()
 
243
 
244
 
245
  ## License
246
+ * The license for the original implementation of SINet can be found
247
+ [here](https://github.com/clovaai/ext_portrait_segmentation/blob/master/LICENSE).
248
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
249
 
250