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
6edd193
·
verified ·
1 Parent(s): e3b5603

v0.46.1

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.46.1 for changelog.

Files changed (10) hide show
  1. Lightweight-Face-Detection_float.dlc +0 -3
  2. Lightweight-Face-Detection_float.onnx.zip +0 -3
  3. Lightweight-Face-Detection_float.tflite +0 -3
  4. Lightweight-Face-Detection_w8a16.dlc +0 -3
  5. Lightweight-Face-Detection_w8a16.onnx.zip +0 -3
  6. Lightweight-Face-Detection_w8a8.dlc +0 -3
  7. Lightweight-Face-Detection_w8a8.onnx.zip +0 -3
  8. Lightweight-Face-Detection_w8a8.tflite +0 -3
  9. README.md +161 -280
  10. tool-versions.yaml +0 -4
Lightweight-Face-Detection_float.dlc DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:befbb92f96351d36042007c635c326131e9c9a5359c03d29ab4dc84c61334352
3
- size 3647084
 
 
 
 
Lightweight-Face-Detection_float.onnx.zip DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:7f37c4db8deb26d3f2e2829ddc33312063f65be2ef4985cee75f05f44c6a366b
3
- size 3282783
 
 
 
 
Lightweight-Face-Detection_float.tflite DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:7bd6be1dd34a138df560e9d9eec4138bab0290837a2477f638af4212dc94e72d
3
- size 3538208
 
 
 
 
Lightweight-Face-Detection_w8a16.dlc DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c503676b903f4a7b2037003065e9e46e21266805277ec8741515a6fc8b59d6e
3
- size 1179100
 
 
 
 
Lightweight-Face-Detection_w8a16.onnx.zip DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:3c8fe1cbc6a6ca91756d4e9fe3dd180542991d05e76b8a660fb3134086338876
3
- size 1750522
 
 
 
 
Lightweight-Face-Detection_w8a8.dlc DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:db541bc9c0c7b1943edb7e338f4ce821b4aa333a039f39c99d15e8d3f8f1f93d
3
- size 1179100
 
 
 
 
Lightweight-Face-Detection_w8a8.onnx.zip DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c19e7c09274a758765b56809943d6624d38ec93f28fe166e7e35b999881ba257
3
- size 1753196
 
 
 
 
Lightweight-Face-Detection_w8a8.tflite DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7b85cb7a0958910b5b02b30d9cafd3e132826acdab75c5feea907815dbb3711
3
- size 989808
 
 
 
 
README.md CHANGED
@@ -11,286 +11,170 @@ pipeline_tag: object-detection
11
 
12
  ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/web-assets/model_demo.png)
13
 
14
- # Lightweight-Face-Detection: Optimized for Mobile Deployment
15
- ## Lightweight and efficient face detector
16
 
17
  A small and accurate model for detecting bounding boxes for faces in images. This model's architecture was developed by Qualcomm. The model was trained by Qualcomm on a proprietary dataset of faces, but can be used on any image.
18
 
19
- This repository provides scripts to run Lightweight-Face-Detection on Qualcomm® devices.
20
- More details on model performance across various devices, can be found
21
- [here](https://aihub.qualcomm.com/models/face_det_lite).
22
-
23
-
24
-
25
- ### Model Details
26
-
27
- - **Model Type:** Model_use_case.object_detection
28
- - **Model Stats:**
29
- - Model checkpoint: qfd360_sl_model.pt
30
- - Inference latency: RealTime
31
- - Input resolution: 480x640
32
- - Number of parameters: 878K
33
- - Model size (float): 3.37 MB
34
- - Model size (w8a8): 965 KB
35
- - Model size (w8a16): 1.09 MB
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 | 6.617 ms | 0 - 121 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_DLC | 6.629 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
41
- | Lightweight-Face-Detection | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 4.107 ms | 0 - 140 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_DLC | 4.101 ms | 1 - 146 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
43
- | Lightweight-Face-Detection | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 2.551 ms | 0 - 3 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_DLC | 2.52 ms | 1 - 3 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
45
- | Lightweight-Face-Detection | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 2.671 ms | 0 - 3 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx.zip) |
46
- | Lightweight-Face-Detection | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 3.229 ms | 0 - 121 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
47
- | Lightweight-Face-Detection | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 3.214 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
48
- | Lightweight-Face-Detection | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 6.617 ms | 0 - 121 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
49
- | Lightweight-Face-Detection | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 6.629 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
50
- | Lightweight-Face-Detection | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 3.723 ms | 0 - 129 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
51
- | Lightweight-Face-Detection | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 3.613 ms | 0 - 128 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
52
- | Lightweight-Face-Detection | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 3.229 ms | 0 - 121 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
53
- | Lightweight-Face-Detection | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 3.214 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
54
- | Lightweight-Face-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.564 ms | 0 - 146 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
55
- | Lightweight-Face-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 1.565 ms | 1 - 144 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
56
- | Lightweight-Face-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1.341 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx.zip) |
57
- | Lightweight-Face-Detection | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 1.223 ms | 0 - 126 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 S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 1.197 ms | 1 - 127 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
59
- | Lightweight-Face-Detection | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 1.209 ms | 1 - 100 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx.zip) |
60
- | Lightweight-Face-Detection | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 1.126 ms | 0 - 128 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.tflite) |
61
- | Lightweight-Face-Detection | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 1.108 ms | 0 - 126 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
62
- | Lightweight-Face-Detection | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 1.267 ms | 0 - 103 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx.zip) |
63
- | Lightweight-Face-Detection | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 2.792 ms | 1 - 1 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.dlc) |
64
- | Lightweight-Face-Detection | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.473 ms | 1 - 1 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection.onnx.zip) |
65
- | Lightweight-Face-Detection | w8a16 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | QNN_DLC | 10.977 ms | 1 - 128 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
66
- | Lightweight-Face-Detection | w8a16 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | ONNX | 44.042 ms | 28 - 41 MB | CPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
67
- | Lightweight-Face-Detection | w8a16 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | QNN_DLC | 5.0 ms | 0 - 3 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
68
- | Lightweight-Face-Detection | w8a16 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | ONNX | 86.915 ms | 28 - 32 MB | CPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
69
- | Lightweight-Face-Detection | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 3.07 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
70
- | Lightweight-Face-Detection | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 2.384 ms | 1 - 142 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
71
- | Lightweight-Face-Detection | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 1.7 ms | 1 - 3 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
72
- | Lightweight-Face-Detection | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 2.009 ms | 0 - 3 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
73
- | Lightweight-Face-Detection | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 2.044 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
74
- | Lightweight-Face-Detection | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 3.07 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
75
- | Lightweight-Face-Detection | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 2.288 ms | 0 - 127 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
76
- | Lightweight-Face-Detection | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 2.044 ms | 1 - 121 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
77
- | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.974 ms | 1 - 144 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
78
- | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1.13 ms | 0 - 117 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
79
- | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.821 ms | 1 - 125 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
80
- | Lightweight-Face-Detection | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.946 ms | 0 - 102 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
81
- | Lightweight-Face-Detection | w8a16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 2.919 ms | 1 - 126 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
82
- | Lightweight-Face-Detection | w8a16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 41.719 ms | 29 - 44 MB | CPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
83
- | Lightweight-Face-Detection | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 0.721 ms | 1 - 125 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
84
- | Lightweight-Face-Detection | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 0.927 ms | 0 - 102 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
85
- | Lightweight-Face-Detection | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 1.933 ms | 1 - 1 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.dlc) |
86
- | Lightweight-Face-Detection | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.972 ms | 1 - 1 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a16.onnx.zip) |
87
- | Lightweight-Face-Detection | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | TFLITE | 2.751 ms | 0 - 124 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
88
- | Lightweight-Face-Detection | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | QNN_DLC | 2.747 ms | 0 - 125 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
89
- | Lightweight-Face-Detection | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | ONNX | 12.63 ms | 12 - 28 MB | CPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
90
- | Lightweight-Face-Detection | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | TFLITE | 1.319 ms | 0 - 3 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
91
- | Lightweight-Face-Detection | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | QNN_DLC | 1.295 ms | 0 - 2 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
92
- | Lightweight-Face-Detection | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | ONNX | 17.27 ms | 11 - 16 MB | CPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
93
- | Lightweight-Face-Detection | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 1.106 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
94
- | Lightweight-Face-Detection | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.103 ms | 0 - 120 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
95
- | Lightweight-Face-Detection | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.514 ms | 0 - 138 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
96
- | Lightweight-Face-Detection | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.492 ms | 0 - 138 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
97
- | Lightweight-Face-Detection | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.413 ms | 0 - 3 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
98
- | Lightweight-Face-Detection | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.417 ms | 0 - 3 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
99
- | Lightweight-Face-Detection | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 0.665 ms | 0 - 4 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
100
- | Lightweight-Face-Detection | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 2.388 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
101
- | Lightweight-Face-Detection | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 0.624 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
102
- | Lightweight-Face-Detection | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 1.106 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
103
- | Lightweight-Face-Detection | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.103 ms | 0 - 120 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
104
- | Lightweight-Face-Detection | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 0.833 ms | 0 - 125 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
105
- | Lightweight-Face-Detection | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 0.82 ms | 0 - 126 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
106
- | Lightweight-Face-Detection | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 2.388 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
107
- | Lightweight-Face-Detection | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 0.624 ms | 0 - 119 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
108
- | Lightweight-Face-Detection | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.281 ms | 0 - 136 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
109
- | Lightweight-Face-Detection | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.277 ms | 0 - 138 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
110
- | Lightweight-Face-Detection | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.436 ms | 0 - 118 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
111
- | Lightweight-Face-Detection | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.235 ms | 0 - 122 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
112
- | Lightweight-Face-Detection | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.23 ms | 0 - 125 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
113
- | Lightweight-Face-Detection | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.35 ms | 0 - 101 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
114
- | Lightweight-Face-Detection | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 0.479 ms | 0 - 125 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
115
- | Lightweight-Face-Detection | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.475 ms | 0 - 124 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
116
- | Lightweight-Face-Detection | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 11.53 ms | 13 - 31 MB | CPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
117
- | Lightweight-Face-Detection | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 0.195 ms | 0 - 121 MB | NPU | [Lightweight-Face-Detection.tflite](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.tflite) |
118
- | Lightweight-Face-Detection | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 0.195 ms | 0 - 122 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
119
- | Lightweight-Face-Detection | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 0.314 ms | 0 - 101 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
120
- | Lightweight-Face-Detection | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.544 ms | 0 - 0 MB | NPU | [Lightweight-Face-Detection.dlc](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.dlc) |
121
- | Lightweight-Face-Detection | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.595 ms | 0 - 0 MB | NPU | [Lightweight-Face-Detection.onnx.zip](https://huggingface.co/qualcomm/Lightweight-Face-Detection/blob/main/Lightweight-Face-Detection_w8a8.onnx.zip) |
122
-
123
-
124
-
125
-
126
- ## Installation
127
-
128
-
129
- Install the package via pip:
130
- ```bash
131
- # NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
132
- pip install "qai-hub-models[face-det-lite]"
133
- ```
134
-
135
-
136
- ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
137
-
138
- Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
139
- Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
140
-
141
- With this API token, you can configure your client to run models on the cloud
142
- hosted devices.
143
- ```bash
144
- qai-hub configure --api_token API_TOKEN
145
- ```
146
- Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
147
-
148
-
149
-
150
- ## Demo off target
151
-
152
- The package contains a simple end-to-end demo that downloads pre-trained
153
- weights and runs this model on a sample input.
154
-
155
- ```bash
156
- python -m qai_hub_models.models.face_det_lite.demo
157
- ```
158
-
159
- The above demo runs a reference implementation of pre-processing, model
160
- inference, and post processing.
161
-
162
- **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
163
- environment, please add the following to your cell (instead of the above).
164
- ```
165
- %run -m qai_hub_models.models.face_det_lite.demo
166
- ```
167
-
168
-
169
- ### Run model on a cloud-hosted device
170
-
171
- In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
172
- device. This script does the following:
173
- * Performance check on-device on a cloud-hosted device
174
- * Downloads compiled assets that can be deployed on-device for Android.
175
- * Accuracy check between PyTorch and on-device outputs.
176
-
177
- ```bash
178
- python -m qai_hub_models.models.face_det_lite.export
179
- ```
180
-
181
-
182
-
183
- ## How does this work?
184
-
185
- This [export script](https://aihub.qualcomm.com/models/face_det_lite/qai_hub_models/models/Lightweight-Face-Detection/export.py)
186
- leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
187
- on-device. Lets go through each step below in detail:
188
-
189
- Step 1: **Compile model for on-device deployment**
190
-
191
- To compile a PyTorch model for on-device deployment, we first trace the model
192
- in memory using the `jit.trace` and then call the `submit_compile_job` API.
193
-
194
- ```python
195
- import torch
196
-
197
- import qai_hub as hub
198
- from qai_hub_models.models.face_det_lite import Model
199
-
200
- # Load the model
201
- torch_model = Model.from_pretrained()
202
-
203
- # Device
204
- device = hub.Device("Samsung Galaxy S25")
205
-
206
- # Trace model
207
- input_shape = torch_model.get_input_spec()
208
- sample_inputs = torch_model.sample_inputs()
209
-
210
- pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
211
-
212
- # Compile model on a specific device
213
- compile_job = hub.submit_compile_job(
214
- model=pt_model,
215
- device=device,
216
- input_specs=torch_model.get_input_spec(),
217
- )
218
-
219
- # Get target model to run on-device
220
- target_model = compile_job.get_target_model()
221
-
222
- ```
223
-
224
-
225
- Step 2: **Performance profiling on cloud-hosted device**
226
-
227
- After compiling models from step 1. Models can be profiled model on-device using the
228
- `target_model`. Note that this scripts runs the model on a device automatically
229
- provisioned in the cloud. Once the job is submitted, you can navigate to a
230
- provided job URL to view a variety of on-device performance metrics.
231
- ```python
232
- profile_job = hub.submit_profile_job(
233
- model=target_model,
234
- device=device,
235
- )
236
-
237
- ```
238
-
239
- Step 3: **Verify on-device accuracy**
240
-
241
- To verify the accuracy of the model on-device, you can run on-device inference
242
- on sample input data on the same cloud hosted device.
243
- ```python
244
- input_data = torch_model.sample_inputs()
245
- inference_job = hub.submit_inference_job(
246
- model=target_model,
247
- device=device,
248
- inputs=input_data,
249
- )
250
- on_device_output = inference_job.download_output_data()
251
-
252
- ```
253
- With the output of the model, you can compute like PSNR, relative errors or
254
- spot check the output with expected output.
255
-
256
- **Note**: This on-device profiling and inference requires access to Qualcomm®
257
- AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
258
-
259
-
260
-
261
- ## Run demo on a cloud-hosted device
262
-
263
- You can also run the demo on-device.
264
-
265
- ```bash
266
- python -m qai_hub_models.models.face_det_lite.demo --eval-mode on-device
267
- ```
268
-
269
- **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
270
- environment, please add the following to your cell (instead of the above).
271
- ```
272
- %run -m qai_hub_models.models.face_det_lite.demo -- --eval-mode on-device
273
- ```
274
-
275
-
276
- ## Deploying compiled model to Android
277
-
278
-
279
- The models can be deployed using multiple runtimes:
280
- - TensorFlow Lite (`.tflite` export): [This
281
- tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
282
- guide to deploy the .tflite model in an Android application.
283
-
284
-
285
- - QNN (`.so` export ): This [sample
286
- app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
287
- provides instructions on how to use the `.so` shared library in an Android application.
288
-
289
-
290
- ## View on Qualcomm® AI Hub
291
- Get more details on Lightweight-Face-Detection's performance across various devices [here](https://aihub.qualcomm.com/models/face_det_lite).
292
- Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
293
-
294
 
295
  ## License
296
  * The license for the original implementation of Lightweight-Face-Detection can be found
@@ -298,9 +182,6 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
298
 
299
 
300
 
301
-
302
  ## Community
303
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
304
  * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
305
-
306
-
 
11
 
12
  ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/web-assets/model_demo.png)
13
 
14
+ # Lightweight-Face-Detection: Optimized for Qualcomm Devices
 
15
 
16
  A small and accurate model for detecting bounding boxes for faces in images. This model's architecture was developed by Qualcomm. The model was trained by Qualcomm on a proprietary dataset of faces, but can be used on any image.
17
 
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/face_det_lite) 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
+ There are two ways to deploy this model on your device:
24
+
25
+ ### Option 1: Download Pre-Exported Models
26
+
27
+ Below are pre-exported model assets ready for deployment.
28
+
29
+ | Runtime | Precision | Chipset | SDK Versions | Download |
30
+ |---|---|---|---|---|
31
+ | ONNX | float | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-onnx-float.zip)
32
+ | ONNX | w8a16 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-onnx-w8a16.zip)
33
+ | ONNX | w8a16_mixed_int16 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-onnx-w8a16_mixed_int16.zip)
34
+ | ONNX | w8a8 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-onnx-w8a8.zip)
35
+ | QNN_DLC | float | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-qnn_dlc-float.zip)
36
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-qnn_dlc-w8a16.zip)
37
+ | QNN_DLC | w8a16_mixed_int16 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-qnn_dlc-w8a16_mixed_int16.zip)
38
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-qnn_dlc-w8a8.zip)
39
+ | TFLITE | float | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-tflite-float.zip)
40
+ | TFLITE | w8a8 | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/face_det_lite/releases/v0.46.1/face_det_lite-tflite-w8a8.zip)
41
+
42
+ For more device-specific assets and performance metrics, visit **[Lightweight-Face-Detection on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/face_det_lite)**.
43
+
44
+
45
+ ### Option 2: Export with Custom Configurations
46
+
47
+ Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/face_det_lite) Python library to compile and export the model with your own:
48
+ - Custom weights (e.g., fine-tuned checkpoints)
49
+ - Custom input shapes
50
+ - Target device and runtime configurations
51
+
52
+ This option is ideal if you need to customize the model beyond the default configuration provided here.
53
+
54
+ See our repository for [Lightweight-Face-Detection on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/face_det_lite) for usage instructions.
55
+
56
+ ## Model Details
57
+
58
+ **Model Type:** Model_use_case.object_detection
59
+
60
+ **Model Stats:**
61
+ - Model checkpoint: qfd360_sl_model.pt
62
+ - Inference latency: RealTime
63
+ - Input resolution: 480x640
64
+ - Number of parameters: 878K
65
+ - Model size (float): 3.37 MB
66
+ - Model size (w8a8): 965 KB
67
+ - Model size (w8a16): 1.09 MB
68
+
69
+ ## Performance Summary
70
+ | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
71
+ |---|---|---|---|---|---|---
72
+ | Lightweight-Face-Detection | ONNX | float | Snapdragon® X Elite | 2.462 ms | 1 - 1 MB | NPU
73
+ | Lightweight-Face-Detection | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.408 ms | 0 - 112 MB | NPU
74
+ | Lightweight-Face-Detection | ONNX | float | Qualcomm® QCS8550 (Proxy) | 2.706 ms | 0 - 7 MB | NPU
75
+ | Lightweight-Face-Detection | ONNX | float | Qualcomm® QCS9075 | 2.694 ms | 1 - 4 MB | NPU
76
+ | Lightweight-Face-Detection | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.217 ms | 0 - 96 MB | NPU
77
+ | Lightweight-Face-Detection | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.256 ms | 0 - 103 MB | NPU
78
+ | Lightweight-Face-Detection | ONNX | w8a16 | Snapdragon® X Elite | 1.944 ms | 1 - 1 MB | NPU
79
+ | Lightweight-Face-Detection | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.189 ms | 0 - 114 MB | NPU
80
+ | Lightweight-Face-Detection | ONNX | w8a16 | Qualcomm® QCS6490 | 87.108 ms | 29 - 32 MB | CPU
81
+ | Lightweight-Face-Detection | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.935 ms | 0 - 122 MB | NPU
82
+ | Lightweight-Face-Detection | ONNX | w8a16 | Qualcomm® QCS9075 | 2.412 ms | 0 - 3 MB | NPU
83
+ | Lightweight-Face-Detection | ONNX | w8a16 | Qualcomm® QCM6690 | 43.818 ms | 29 - 36 MB | CPU
84
+ | Lightweight-Face-Detection | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.947 ms | 0 - 101 MB | NPU
85
+ | Lightweight-Face-Detection | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 33.473 ms | 29 - 36 MB | CPU
86
+ | Lightweight-Face-Detection | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.919 ms | 0 - 102 MB | NPU
87
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Snapdragon® X Elite | 2.027 ms | 1 - 1 MB | NPU
88
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 1.248 ms | 0 - 111 MB | NPU
89
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Qualcomm® QCS6490 | 91.198 ms | 28 - 32 MB | CPU
90
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Qualcomm® QCS8550 (Proxy) | 2.063 ms | 0 - 8 MB | NPU
91
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Qualcomm® QCS9075 | 2.288 ms | 0 - 3 MB | NPU
92
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Qualcomm® QCM6690 | 43.997 ms | 28 - 35 MB | CPU
93
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.988 ms | 0 - 103 MB | NPU
94
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 33.242 ms | 30 - 37 MB | CPU
95
+ | Lightweight-Face-Detection | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.93 ms | 0 - 101 MB | NPU
96
+ | Lightweight-Face-Detection | ONNX | w8a8 | Snapdragon® X Elite | 0.554 ms | 0 - 0 MB | NPU
97
+ | Lightweight-Face-Detection | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.474 ms | 0 - 107 MB | NPU
98
+ | Lightweight-Face-Detection | ONNX | w8a8 | Qualcomm® QCS6490 | 16.911 ms | 12 - 17 MB | CPU
99
+ | Lightweight-Face-Detection | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.623 ms | 0 - 3 MB | NPU
100
+ | Lightweight-Face-Detection | ONNX | w8a8 | Qualcomm® QCS9075 | 0.838 ms | 0 - 3 MB | NPU
101
+ | Lightweight-Face-Detection | ONNX | w8a8 | Qualcomm® QCM6690 | 12.739 ms | 13 - 21 MB | CPU
102
+ | Lightweight-Face-Detection | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.354 ms | 0 - 100 MB | NPU
103
+ | Lightweight-Face-Detection | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 9.371 ms | 12 - 20 MB | CPU
104
+ | Lightweight-Face-Detection | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.331 ms | 0 - 100 MB | NPU
105
+ | Lightweight-Face-Detection | QNN_DLC | float | Snapdragon® X Elite | 2.872 ms | 1 - 1 MB | NPU
106
+ | Lightweight-Face-Detection | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.621 ms | 0 - 42 MB | NPU
107
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 6.779 ms | 1 - 29 MB | NPU
108
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 2.602 ms | 1 - 24 MB | NPU
109
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® SA8775P | 14.285 ms | 1 - 29 MB | NPU
110
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® QCS9075 | 3.692 ms | 1 - 4 MB | NPU
111
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 4.31 ms | 0 - 43 MB | NPU
112
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® SA7255P | 6.779 ms | 1 - 29 MB | NPU
113
+ | Lightweight-Face-Detection | QNN_DLC | float | Qualcomm® SA8295P | 3.79 ms | 0 - 25 MB | NPU
114
+ | Lightweight-Face-Detection | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.218 ms | 1 - 34 MB | NPU
115
+ | Lightweight-Face-Detection | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.1 ms | 1 - 34 MB | NPU
116
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.909 ms | 1 - 1 MB | NPU
117
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.939 ms | 0 - 155 MB | NPU
118
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 6.589 ms | 2 - 5 MB | NPU
119
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 2.989 ms | 1 - 139 MB | NPU
120
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 1.675 ms | 1 - 2 MB | NPU
121
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.941 ms | 1 - 141 MB | NPU
122
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 2.171 ms | 2 - 5 MB | NPU
123
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 12.4 ms | 1 - 140 MB | NPU
124
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 2.894 ms | 1 - 155 MB | NPU
125
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® SA7255P | 2.989 ms | 1 - 139 MB | NPU
126
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Qualcomm® SA8295P | 2.34 ms | 1 - 137 MB | NPU
127
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.774 ms | 0 - 143 MB | NPU
128
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 2.605 ms | 1 - 139 MB | NPU
129
+ | Lightweight-Face-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.713 ms | 1 - 143 MB | NPU
130
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X Elite | 1.979 ms | 1 - 1 MB | NPU
131
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 0.959 ms | 0 - 153 MB | NPU
132
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCS8275 (Proxy) | 3.246 ms | 1 - 140 MB | NPU
133
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCS8550 (Proxy) | 1.739 ms | 1 - 2 MB | NPU
134
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8775P | 1.957 ms | 0 - 142 MB | NPU
135
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCS9075 | 2.37 ms | 0 - 3 MB | NPU
136
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCM6690 | 14.536 ms | 1 - 140 MB | NPU
137
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA7255P | 3.246 ms | 1 - 140 MB | NPU
138
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.815 ms | 0 - 143 MB | NPU
139
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 2.629 ms | 1 - 139 MB | NPU
140
+ | Lightweight-Face-Detection | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.756 ms | 1 - 143 MB | NPU
141
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.536 ms | 0 - 0 MB | NPU
142
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.276 ms | 0 - 34 MB | NPU
143
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 1.41 ms | 2 - 4 MB | NPU
144
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.081 ms | 0 - 26 MB | NPU
145
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.414 ms | 0 - 2 MB | NPU
146
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.614 ms | 0 - 26 MB | NPU
147
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.553 ms | 0 - 2 MB | NPU
148
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 2.813 ms | 0 - 141 MB | NPU
149
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.499 ms | 0 - 36 MB | NPU
150
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.081 ms | 0 - 26 MB | NPU
151
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.823 ms | 0 - 23 MB | NPU
152
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.23 ms | 0 - 28 MB | NPU
153
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.487 ms | 0 - 138 MB | NPU
154
+ | Lightweight-Face-Detection | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.198 ms | 0 - 27 MB | NPU
155
+ | Lightweight-Face-Detection | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.633 ms | 0 - 43 MB | NPU
156
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 6.711 ms | 0 - 29 MB | NPU
157
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 2.607 ms | 0 - 59 MB | NPU
158
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® SA8775P | 3.346 ms | 0 - 31 MB | NPU
159
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® QCS9075 | 3.709 ms | 0 - 5 MB | NPU
160
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 4.285 ms | 0 - 43 MB | NPU
161
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® SA7255P | 6.711 ms | 0 - 29 MB | NPU
162
+ | Lightweight-Face-Detection | TFLITE | float | Qualcomm® SA8295P | 3.783 ms | 0 - 27 MB | NPU
163
+ | Lightweight-Face-Detection | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.231 ms | 0 - 34 MB | NPU
164
+ | Lightweight-Face-Detection | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.11 ms | 0 - 34 MB | NPU
165
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.284 ms | 0 - 34 MB | NPU
166
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® QCS6490 | 1.412 ms | 0 - 3 MB | NPU
167
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.126 ms | 0 - 24 MB | NPU
168
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.411 ms | 0 - 1 MB | NPU
169
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® SA8775P | 0.623 ms | 0 - 26 MB | NPU
170
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.574 ms | 0 - 3 MB | NPU
171
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® QCM6690 | 2.799 ms | 0 - 136 MB | NPU
172
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.507 ms | 0 - 35 MB | NPU
173
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® SA7255P | 1.126 ms | 0 - 24 MB | NPU
174
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Qualcomm® SA8295P | 0.829 ms | 0 - 22 MB | NPU
175
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.235 ms | 0 - 23 MB | NPU
176
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.496 ms | 0 - 136 MB | NPU
177
+ | Lightweight-Face-Detection | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.197 ms | 0 - 27 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
178
 
179
  ## License
180
  * The license for the original implementation of Lightweight-Face-Detection can be found
 
182
 
183
 
184
 
 
185
  ## Community
186
  * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
187
  * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
 
 
tool-versions.yaml DELETED
@@ -1,4 +0,0 @@
1
- tool_versions:
2
- onnx:
3
- qairt: 2.37.1.250807093845_124904
4
- onnx_runtime: 1.23.0