Hugging Face's logo

qualcomm
/
Lightweight-Face-Detection

Object Detection
PyTorch
real_time
bu_iot
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
dac237c
·
verified ·
1 Parent(s): 94688a0

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +57 -26
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: pytorch
3
- license: bsd-3-clause
4
  tags:
5
  - real_time
6
  - android
@@ -26,32 +26,63 @@ More details on model performance across various devices, can be found
26
 
27
  ### Model Details
28
 
29
- - **Model Type:** Object detection
30
  - **Model Stats:**
31
  - Model checkpoint: qfd360_sl_model.pt
32
  - Inference latency: RealTime
33
  - Input resolution: 480x640
34
  - Number of parameters: 0.89M
35
- - Model size: 3.4M
36
 
37
- | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
- | Lightweight-Face-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.99 ms | 0 - 7 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
40
- | Lightweight-Face-Detection | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.993 ms | 1 - 12 MB | FP16 | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
41
- | Lightweight-Face-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.87 ms | 0 - 23 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
42
- | Lightweight-Face-Detection | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.365 ms | 0 - 34 MB | FP16 | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
43
- | Lightweight-Face-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.56 ms | 0 - 19 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
44
- | Lightweight-Face-Detection | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.58 ms | 0 - 25 MB | FP16 | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
45
- | Lightweight-Face-Detection | SA7255P ADP | SA7255P | TFLITE | 32.511 ms | 0 - 14 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
46
- | Lightweight-Face-Detection | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.987 ms | 0 - 7 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
47
- | Lightweight-Face-Detection | SA8295P ADP | SA8295P | TFLITE | 4.248 ms | 0 - 19 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
48
- | Lightweight-Face-Detection | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.999 ms | 0 - 7 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
49
- | Lightweight-Face-Detection | SA8775P ADP | SA8775P | TFLITE | 4.592 ms | 0 - 14 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
50
- | Lightweight-Face-Detection | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 32.511 ms | 0 - 14 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
51
- | Lightweight-Face-Detection | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 3.01 ms | 0 - 7 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
52
- | Lightweight-Face-Detection | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 4.592 ms | 0 - 14 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
53
- | Lightweight-Face-Detection | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 3.614 ms | 0 - 31 MB | FP16 | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
54
- | Lightweight-Face-Detection | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.073 ms | 1 - 1 MB | FP16 | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55
 
56
 
57
 
@@ -113,12 +144,12 @@ python -m qai_hub_models.models.face_det_lite.export
113
  Profiling Results
114
  ------------------------------------------------------------
115
  Lightweight-Face-Detection
116
- Device : Samsung Galaxy S23 (13)
117
- Runtime : TFLITE
118
- Estimated inference time (ms) : 3.0
119
- Estimated peak memory usage (MB): [0, 7]
120
- Total # Ops : 86
121
- Compute Unit(s) : NPU (86 ops)
122
  ```
123
 
124
 
 
1
  ---
2
  library_name: pytorch
3
+ license: other
4
  tags:
5
  - real_time
6
  - android
 
26
 
27
  ### Model Details
28
 
29
+ - **Model Type:** Model_use_case.object_detection
30
  - **Model Stats:**
31
  - Model checkpoint: qfd360_sl_model.pt
32
  - Inference latency: RealTime
33
  - Input resolution: 480x640
34
  - Number of parameters: 0.89M
35
+ - Model size (float): 3.4M
36
 
37
+ | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
+ | Lightweight-Face-Detection | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 7.656 ms | 0 - 14 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
40
+ | Lightweight-Face-Detection | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 6.485 ms | 1 - 11 MB | NPU | Use Export Script |
41
+ | Lightweight-Face-Detection | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 3.59 ms | 0 - 24 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
42
+ | Lightweight-Face-Detection | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 4.23 ms | 1 - 22 MB | NPU | Use Export Script |
43
+ | Lightweight-Face-Detection | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 2.996 ms | 0 - 7 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
44
+ | Lightweight-Face-Detection | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 2.466 ms | 1 - 3 MB | NPU | Use Export Script |
45
+ | Lightweight-Face-Detection | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 3.759 ms | 0 - 17 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
46
+ | Lightweight-Face-Detection | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 3.093 ms | 1 - 16 MB | NPU | Use Export Script |
47
+ | Lightweight-Face-Detection | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 7.656 ms | 0 - 14 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
48
+ | Lightweight-Face-Detection | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 6.485 ms | 1 - 11 MB | NPU | Use Export Script |
49
+ | Lightweight-Face-Detection | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 2.982 ms | 0 - 6 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
50
+ | Lightweight-Face-Detection | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 2.47 ms | 0 - 2 MB | NPU | Use Export Script |
51
+ | Lightweight-Face-Detection | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 4.229 ms | 0 - 18 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
52
+ | Lightweight-Face-Detection | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 3.576 ms | 0 - 18 MB | NPU | Use Export Script |
53
+ | Lightweight-Face-Detection | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 2.99 ms | 0 - 8 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
54
+ | Lightweight-Face-Detection | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 2.45 ms | 1 - 3 MB | NPU | Use Export Script |
55
+ | Lightweight-Face-Detection | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 3.759 ms | 0 - 17 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
56
+ | Lightweight-Face-Detection | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 3.093 ms | 1 - 16 MB | NPU | Use Export Script |
57
+ | Lightweight-Face-Detection | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 2.982 ms | 0 - 7 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
58
+ | Lightweight-Face-Detection | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 2.462 ms | 1 - 6 MB | NPU | Use Export Script |
59
+ | Lightweight-Face-Detection | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 2.011 ms | 0 - 11 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
60
+ | Lightweight-Face-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.87 ms | 0 - 29 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
61
+ | Lightweight-Face-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 1.583 ms | 0 - 22 MB | NPU | Use Export Script |
62
+ | Lightweight-Face-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1.339 ms | 0 - 34 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
63
+ | Lightweight-Face-Detection | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 1.918 ms | 0 - 20 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
64
+ | Lightweight-Face-Detection | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 1.541 ms | 1 - 23 MB | NPU | Use Export Script |
65
+ | Lightweight-Face-Detection | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 1.29 ms | 0 - 26 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
66
+ | Lightweight-Face-Detection | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.753 ms | 1 - 1 MB | NPU | Use Export Script |
67
+ | Lightweight-Face-Detection | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.109 ms | 0 - 0 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx) |
68
+ | Lightweight-Face-Detection | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 3.203 ms | 0 - 10 MB | NPU | Use Export Script |
69
+ | Lightweight-Face-Detection | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 3.22 ms | 1 - 31 MB | NPU | Use Export Script |
70
+ | Lightweight-Face-Detection | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 1.607 ms | 1 - 3 MB | NPU | Use Export Script |
71
+ | Lightweight-Face-Detection | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 1.892 ms | 1 - 15 MB | NPU | Use Export Script |
72
+ | Lightweight-Face-Detection | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 8.633 ms | 1 - 12 MB | NPU | Use Export Script |
73
+ | Lightweight-Face-Detection | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN | 3.203 ms | 0 - 10 MB | NPU | Use Export Script |
74
+ | Lightweight-Face-Detection | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 1.602 ms | 1 - 3 MB | NPU | Use Export Script |
75
+ | Lightweight-Face-Detection | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN | 2.108 ms | 0 - 18 MB | NPU | Use Export Script |
76
+ | Lightweight-Face-Detection | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 1.605 ms | 1 - 3 MB | NPU | Use Export Script |
77
+ | Lightweight-Face-Detection | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN | 1.892 ms | 1 - 15 MB | NPU | Use Export Script |
78
+ | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 1.609 ms | 1 - 6 MB | NPU | Use Export Script |
79
+ | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 1.98 ms | 0 - 9 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx) |
80
+ | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 0.97 ms | 1 - 30 MB | NPU | Use Export Script |
81
+ | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1.194 ms | 0 - 33 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx) |
82
+ | Lightweight-Face-Detection | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 1.034 ms | 1 - 24 MB | NPU | Use Export Script |
83
+ | Lightweight-Face-Detection | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 0.906 ms | 0 - 26 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx) |
84
+ | Lightweight-Face-Detection | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.818 ms | 1 - 1 MB | NPU | Use Export Script |
85
+ | Lightweight-Face-Detection | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.093 ms | 1 - 1 MB | NPU | [Lightweight-Face-Detection.onnx](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx) |
86
 
87
 
88
 
 
144
  Profiling Results
145
  ------------------------------------------------------------
146
  Lightweight-Face-Detection
147
+ Device : cs_8275 (ANDROID 14)
148
+ Runtime : TFLITE
149
+ Estimated inference time (ms) : 7.7
150
+ Estimated peak memory usage (MB): [0, 14]
151
+ Total # Ops : 86
152
+ Compute Unit(s) : npu (86 ops) gpu (0 ops) cpu (0 ops)
153
  ```
154
 
155