Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
library_name: pytorch
|
| 3 |
-
license:
|
| 4 |
tags:
|
| 5 |
- android
|
| 6 |
pipeline_tag: image-segmentation
|
|
@@ -25,7 +25,7 @@ More details on model performance across various devices, can be found
|
|
| 25 |
|
| 26 |
### Model Details
|
| 27 |
|
| 28 |
-
- **Model Type:**
|
| 29 |
- **Model Stats:**
|
| 30 |
- Model checkpoint: Square
|
| 31 |
- Input resolution (Square): 256x256
|
|
@@ -34,37 +34,37 @@ More details on model performance across various devices, can be found
|
|
| 34 |
- Model size: 454 KB
|
| 35 |
- Number of output classes: 6
|
| 36 |
|
| 37 |
-
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) |
|
| 38 |
|---|---|---|---|---|---|---|---|---|
|
| 39 |
-
| MediaPipe-Selfie-Segmentation |
|
| 40 |
-
| MediaPipe-Selfie-Segmentation |
|
| 41 |
-
| MediaPipe-Selfie-Segmentation |
|
| 42 |
-
| MediaPipe-Selfie-Segmentation |
|
| 43 |
-
| MediaPipe-Selfie-Segmentation |
|
| 44 |
-
| MediaPipe-Selfie-Segmentation |
|
| 45 |
-
| MediaPipe-Selfie-Segmentation |
|
| 46 |
-
| MediaPipe-Selfie-Segmentation |
|
| 47 |
-
| MediaPipe-Selfie-Segmentation |
|
| 48 |
-
| MediaPipe-Selfie-Segmentation | SA7255P ADP | SA7255P |
|
| 49 |
-
| MediaPipe-Selfie-Segmentation |
|
| 50 |
-
| MediaPipe-Selfie-Segmentation | SA8255 (Proxy) | SA8255P Proxy |
|
| 51 |
-
| MediaPipe-Selfie-Segmentation |
|
| 52 |
-
| MediaPipe-Selfie-Segmentation | SA8295P ADP | SA8295P |
|
| 53 |
-
| MediaPipe-Selfie-Segmentation |
|
| 54 |
-
| MediaPipe-Selfie-Segmentation | SA8650 (Proxy) | SA8650P Proxy |
|
| 55 |
-
| MediaPipe-Selfie-Segmentation |
|
| 56 |
-
| MediaPipe-Selfie-Segmentation | SA8775P ADP | SA8775P |
|
| 57 |
-
| MediaPipe-Selfie-Segmentation |
|
| 58 |
-
| MediaPipe-Selfie-Segmentation |
|
| 59 |
-
| MediaPipe-Selfie-Segmentation |
|
| 60 |
-
| MediaPipe-Selfie-Segmentation |
|
| 61 |
-
| MediaPipe-Selfie-Segmentation |
|
| 62 |
-
| MediaPipe-Selfie-Segmentation |
|
| 63 |
-
| MediaPipe-Selfie-Segmentation |
|
| 64 |
-
| MediaPipe-Selfie-Segmentation |
|
| 65 |
-
| MediaPipe-Selfie-Segmentation |
|
| 66 |
-
| MediaPipe-Selfie-Segmentation | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.
|
| 67 |
-
| MediaPipe-Selfie-Segmentation | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.
|
| 68 |
|
| 69 |
|
| 70 |
|
|
@@ -126,12 +126,12 @@ python -m qai_hub_models.models.mediapipe_selfie.export
|
|
| 126 |
Profiling Results
|
| 127 |
------------------------------------------------------------
|
| 128 |
MediaPipe-Selfie-Segmentation
|
| 129 |
-
Device :
|
| 130 |
-
Runtime : TFLITE
|
| 131 |
-
Estimated inference time (ms) :
|
| 132 |
-
Estimated peak memory usage (MB): [0,
|
| 133 |
-
Total # Ops : 118
|
| 134 |
-
Compute Unit(s) :
|
| 135 |
```
|
| 136 |
|
| 137 |
|
|
|
|
| 1 |
---
|
| 2 |
library_name: pytorch
|
| 3 |
+
license: other
|
| 4 |
tags:
|
| 5 |
- android
|
| 6 |
pipeline_tag: image-segmentation
|
|
|
|
| 25 |
|
| 26 |
### Model Details
|
| 27 |
|
| 28 |
+
- **Model Type:** Model_use_case.semantic_segmentation
|
| 29 |
- **Model Stats:**
|
| 30 |
- Model checkpoint: Square
|
| 31 |
- Input resolution (Square): 256x256
|
|
|
|
| 34 |
- Model size: 454 KB
|
| 35 |
- Number of output classes: 6
|
| 36 |
|
| 37 |
+
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|
| 38 |
|---|---|---|---|---|---|---|---|---|
|
| 39 |
+
| MediaPipe-Selfie-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 1.855 ms | 0 - 12 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 40 |
+
| MediaPipe-Selfie-Segmentation | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 1.749 ms | 1 - 10 MB | NPU | Use Export Script |
|
| 41 |
+
| MediaPipe-Selfie-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.993 ms | 0 - 20 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 42 |
+
| MediaPipe-Selfie-Segmentation | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 1.056 ms | 1 - 23 MB | NPU | Use Export Script |
|
| 43 |
+
| MediaPipe-Selfie-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.694 ms | 0 - 8 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 44 |
+
| MediaPipe-Selfie-Segmentation | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 0.692 ms | 1 - 3 MB | NPU | Use Export Script |
|
| 45 |
+
| MediaPipe-Selfie-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 1.025 ms | 0 - 15 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 46 |
+
| MediaPipe-Selfie-Segmentation | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 0.978 ms | 1 - 16 MB | NPU | Use Export Script |
|
| 47 |
+
| MediaPipe-Selfie-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 1.855 ms | 0 - 12 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 48 |
+
| MediaPipe-Selfie-Segmentation | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 1.749 ms | 1 - 10 MB | NPU | Use Export Script |
|
| 49 |
+
| MediaPipe-Selfie-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.692 ms | 0 - 8 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 50 |
+
| MediaPipe-Selfie-Segmentation | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 0.704 ms | 1 - 3 MB | NPU | Use Export Script |
|
| 51 |
+
| MediaPipe-Selfie-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.273 ms | 0 - 16 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 52 |
+
| MediaPipe-Selfie-Segmentation | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 1.265 ms | 0 - 6 MB | NPU | Use Export Script |
|
| 53 |
+
| MediaPipe-Selfie-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.693 ms | 0 - 7 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 54 |
+
| MediaPipe-Selfie-Segmentation | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 0.696 ms | 2 - 4 MB | NPU | Use Export Script |
|
| 55 |
+
| MediaPipe-Selfie-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 1.025 ms | 0 - 15 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 56 |
+
| MediaPipe-Selfie-Segmentation | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 0.978 ms | 1 - 16 MB | NPU | Use Export Script |
|
| 57 |
+
| MediaPipe-Selfie-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 0.692 ms | 0 - 7 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 58 |
+
| MediaPipe-Selfie-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 0.691 ms | 1 - 9 MB | NPU | Use Export Script |
|
| 59 |
+
| MediaPipe-Selfie-Segmentation | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 0.976 ms | 0 - 6 MB | NPU | [MediaPipe-Selfie-Segmentation.onnx](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.onnx) |
|
| 60 |
+
| MediaPipe-Selfie-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.455 ms | 0 - 27 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 61 |
+
| MediaPipe-Selfie-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 0.457 ms | 0 - 26 MB | NPU | Use Export Script |
|
| 62 |
+
| MediaPipe-Selfie-Segmentation | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.64 ms | 0 - 30 MB | NPU | [MediaPipe-Selfie-Segmentation.onnx](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.onnx) |
|
| 63 |
+
| MediaPipe-Selfie-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 0.43 ms | 0 - 18 MB | NPU | [MediaPipe-Selfie-Segmentation.tflite](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.tflite) |
|
| 64 |
+
| MediaPipe-Selfie-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 0.438 ms | 0 - 18 MB | NPU | Use Export Script |
|
| 65 |
+
| MediaPipe-Selfie-Segmentation | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 0.628 ms | 1 - 21 MB | NPU | [MediaPipe-Selfie-Segmentation.onnx](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.onnx) |
|
| 66 |
+
| MediaPipe-Selfie-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.835 ms | 1 - 1 MB | NPU | Use Export Script |
|
| 67 |
+
| MediaPipe-Selfie-Segmentation | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.079 ms | 2 - 2 MB | NPU | [MediaPipe-Selfie-Segmentation.onnx](https://huggingface.co/qualcomm/MediaPipe-Selfie-Segmentation/blob/main/MediaPipe-Selfie-Segmentation.onnx) |
|
| 68 |
|
| 69 |
|
| 70 |
|
|
|
|
| 126 |
Profiling Results
|
| 127 |
------------------------------------------------------------
|
| 128 |
MediaPipe-Selfie-Segmentation
|
| 129 |
+
Device : cs_8275 (ANDROID 14)
|
| 130 |
+
Runtime : TFLITE
|
| 131 |
+
Estimated inference time (ms) : 1.9
|
| 132 |
+
Estimated peak memory usage (MB): [0, 12]
|
| 133 |
+
Total # Ops : 118
|
| 134 |
+
Compute Unit(s) : npu (118 ops) gpu (0 ops) cpu (0 ops)
|
| 135 |
```
|
| 136 |
|
| 137 |
|