Hugging Face's logo

aplux
/
Facial-Landmark-Detection

Object Detection
AIoT
QNN
Model card Files Community
qc903113684 commited on
Commit
b316dda
·
verified ·
1 Parent(s): aaa4363

Upload 15 files

Browse files
Files changed (15) hide show
  1. model_farm_facial_landmark_det_qcs6490_qnn2.31_int8_aidlite/README.md +40 -0
  2. model_farm_facial_landmark_det_qcs6490_qnn2.31_int8_aidlite/models/facial_landmark_det_w8a8.qnn231.ctx.bin +3 -0
  3. model_farm_facial_landmark_det_qcs6490_qnn2.31_int8_aidlite/python/test.jpg +0 -0
  4. model_farm_facial_landmark_det_qcs6490_qnn2.31_w8a16_aidlite/README.md +40 -0
  5. model_farm_facial_landmark_det_qcs6490_qnn2.31_w8a16_aidlite/models/facial_landmark_det_w8a16.qnn231.ctx.bin +3 -0
  6. model_farm_facial_landmark_det_qcs6490_qnn2.31_w8a16_aidlite/python/test.jpg +0 -0
  7. model_farm_facial_landmark_det_qcs8550_qnn2.31_fp16_aidlite/README.md +40 -0
  8. model_farm_facial_landmark_det_qcs8550_qnn2.31_fp16_aidlite/models/facial_landmark_det_fp16.qnn231.ctx.bin +3 -0
  9. model_farm_facial_landmark_det_qcs8550_qnn2.31_fp16_aidlite/python/test.jpg +0 -0
  10. model_farm_facial_landmark_det_qcs8550_qnn2.31_int8_aidlite/README.md +40 -0
  11. model_farm_facial_landmark_det_qcs8550_qnn2.31_int8_aidlite/models/facial_landmark_det_w8a8.qnn231.ctx.bin +3 -0
  12. model_farm_facial_landmark_det_qcs8550_qnn2.31_int8_aidlite/python/test.jpg +0 -0
  13. model_farm_facial_landmark_det_qcs8550_qnn2.31_w8a16_aidlite/README.md +40 -0
  14. model_farm_facial_landmark_det_qcs8550_qnn2.31_w8a16_aidlite/converted_models/facial_landmark_det_w8a16.qnn231.ctx.bin +3 -0
  15. model_farm_facial_landmark_det_qcs8550_qnn2.31_w8a16_aidlite/python/test.jpg +0 -0
model_farm_facial_landmark_det_qcs6490_qnn2.31_int8_aidlite/README.md ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## Model Information
2
+
3
+ ### Source model
4
+ - Input shape: [1x3x128x128]
5
+ - Number of parameters: 5.17M
6
+ - Model size: 20.95M
7
+ - Output shape: [1x265]
8
+
9
+ Source model repository: [Facial-Landmark-Detection](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/facemap_3dmm/model.py)
10
+
11
+ ### Converted model
12
+
13
+ - Precision: INT8
14
+ - Backend: QNN2.31
15
+ - Target Device: FV01 QCS6490
16
+
17
+ ## Inference with AidLite SDK
18
+
19
+ ### SDK installation
20
+ Model Farm uses AidLite SDK as the model inference SDK. For details, please refer to the [AidLite Developer Documentation](https://v2.docs.aidlux.com/en/sdk-api/aidlite-sdk/)
21
+
22
+ - Install AidLite SDK
23
+
24
+ ```bash
25
+ # Install the appropriate version of the aidlite sdk
26
+ sudo aid-pkg update
27
+ sudo aid-pkg install aidlite-sdk
28
+ # Download the qnn version that matches the above backend. Eg Install QNN2.23 Aidlite: sudo aid-pkg install aidlite-qnn223
29
+ sudo aid-pkg install aidlite-{QNN VERSION}
30
+ ```
31
+
32
+ - Verify AidLite SDK
33
+
34
+ ```bash
35
+ # aidlite sdk c++ check
36
+ python3 -c "import aidlite ; print(aidlite.get_library_version())"
37
+
38
+ # aidlite sdk python check
39
+ python3 -c "import aidlite ; print(aidlite.get_py_library_version())"
40
+ ```
model_farm_facial_landmark_det_qcs6490_qnn2.31_int8_aidlite/models/facial_landmark_det_w8a8.qnn231.ctx.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03c8b5a130676bfc65932238916464229b0363e2229fc2ab4d5388e239635f65
3
+ size 5515808
model_farm_facial_landmark_det_qcs6490_qnn2.31_int8_aidlite/python/test.jpg ADDED
model_farm_facial_landmark_det_qcs6490_qnn2.31_w8a16_aidlite/README.md ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## Model Information
2
+
3
+ ### Source model
4
+ - Input shape: [1x3x128x128]
5
+ - Number of parameters: 5.17M
6
+ - Model size: 20.95M
7
+ - Output shape: [1x265]
8
+
9
+ Source model repository: [Facial-Landmark-Detection](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/facemap_3dmm/model.py)
10
+
11
+ ### Converted model
12
+
13
+ - Precision: W8A16
14
+ - Backend: QNN2.31
15
+ - Target Device: FV01 QCS6490
16
+
17
+ ## Inference with AidLite SDK
18
+
19
+ ### SDK installation
20
+ Model Farm uses AidLite SDK as the model inference SDK. For details, please refer to the [AidLite Developer Documentation](https://v2.docs.aidlux.com/en/sdk-api/aidlite-sdk/)
21
+
22
+ - Install AidLite SDK
23
+
24
+ ```bash
25
+ # Install the appropriate version of the aidlite sdk
26
+ sudo aid-pkg update
27
+ sudo aid-pkg install aidlite-sdk
28
+ # Download the qnn version that matches the above backend. Eg Install QNN2.23 Aidlite: sudo aid-pkg install aidlite-qnn223
29
+ sudo aid-pkg install aidlite-{QNN VERSION}
30
+ ```
31
+
32
+ - Verify AidLite SDK
33
+
34
+ ```bash
35
+ # aidlite sdk c++ check
36
+ python3 -c "import aidlite ; print(aidlite.get_library_version())"
37
+
38
+ # aidlite sdk python check
39
+ python3 -c "import aidlite ; print(aidlite.get_py_library_version())"
40
+ ```
model_farm_facial_landmark_det_qcs6490_qnn2.31_w8a16_aidlite/models/facial_landmark_det_w8a16.qnn231.ctx.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a42b686b432dd75ae9c74b2083542fe2c44e28016966f4147b969a67c5e42fc
3
+ size 5519904
model_farm_facial_landmark_det_qcs6490_qnn2.31_w8a16_aidlite/python/test.jpg ADDED
model_farm_facial_landmark_det_qcs8550_qnn2.31_fp16_aidlite/README.md ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## Model Information
2
+
3
+ ### Source model
4
+ - Input shape: [1x3x128x128]
5
+ - Number of parameters: 5.17M
6
+ - Model size: 20.95M
7
+ - Output shape: [1x265]
8
+
9
+ Source model repository: [Facial-Landmark-Detection](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/facemap_3dmm/model.py)
10
+
11
+ ### Converted model
12
+
13
+ - Precision: FP16
14
+ - Backend: QNN2.31
15
+ - Target Device: SNM972 QCS8550
16
+
17
+ ## Inference with AidLite SDK
18
+
19
+ ### SDK installation
20
+ Model Farm uses AidLite SDK as the model inference SDK. For details, please refer to the [AidLite Developer Documentation](https://v2.docs.aidlux.com/en/sdk-api/aidlite-sdk/)
21
+
22
+ - Install AidLite SDK
23
+
24
+ ```bash
25
+ # Install the appropriate version of the aidlite sdk
26
+ sudo aid-pkg update
27
+ sudo aid-pkg install aidlite-sdk
28
+ # Download the qnn version that matches the above backend. Eg Install QNN2.23 Aidlite: sudo aid-pkg install aidlite-qnn223
29
+ sudo aid-pkg install aidlite-{QNN VERSION}
30
+ ```
31
+
32
+ - Verify AidLite SDK
33
+
34
+ ```bash
35
+ # aidlite sdk c++ check
36
+ python3 -c "import aidlite ; print(aidlite.get_library_version())"
37
+
38
+ # aidlite sdk python check
39
+ python3 -c "import aidlite ; print(aidlite.get_py_library_version())"
40
+ ```
model_farm_facial_landmark_det_qcs8550_qnn2.31_fp16_aidlite/models/facial_landmark_det_fp16.qnn231.ctx.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fb86293fa5423733133a7c08b2655350cd1e49dfb787deba8dc10bf23b45e96
3
+ size 11042416
model_farm_facial_landmark_det_qcs8550_qnn2.31_fp16_aidlite/python/test.jpg ADDED
model_farm_facial_landmark_det_qcs8550_qnn2.31_int8_aidlite/README.md ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## Model Information
2
+
3
+ ### Source model
4
+ - Input shape: [1x3x128x128]
5
+ - Number of parameters: 5.17M
6
+ - Model size: 20.95M
7
+ - Output shape: [1x265]
8
+
9
+ Source model repository: [Facial-Landmark-Detection](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/facemap_3dmm/model.py)
10
+
11
+ ### Converted model
12
+
13
+ - Precision: INT8
14
+ - Backend: QNN2.31
15
+ - Target Device: SNM972 QCS8550
16
+
17
+ ## Inference with AidLite SDK
18
+
19
+ ### SDK installation
20
+ Model Farm uses AidLite SDK as the model inference SDK. For details, please refer to the [AidLite Developer Documentation](https://v2.docs.aidlux.com/en/sdk-api/aidlite-sdk/)
21
+
22
+ - Install AidLite SDK
23
+
24
+ ```bash
25
+ # Install the appropriate version of the aidlite sdk
26
+ sudo aid-pkg update
27
+ sudo aid-pkg install aidlite-sdk
28
+ # Download the qnn version that matches the above backend. Eg Install QNN2.23 Aidlite: sudo aid-pkg install aidlite-qnn223
29
+ sudo aid-pkg install aidlite-{QNN VERSION}
30
+ ```
31
+
32
+ - Verify AidLite SDK
33
+
34
+ ```bash
35
+ # aidlite sdk c++ check
36
+ python3 -c "import aidlite ; print(aidlite.get_library_version())"
37
+
38
+ # aidlite sdk python check
39
+ python3 -c "import aidlite ; print(aidlite.get_py_library_version())"
40
+ ```
model_farm_facial_landmark_det_qcs8550_qnn2.31_int8_aidlite/models/facial_landmark_det_w8a8.qnn231.ctx.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ea3df6f2dc8f72c9ada4ca3af78abb02fe724508235a02b5f8d78730ec7cf96
3
+ size 5515808
model_farm_facial_landmark_det_qcs8550_qnn2.31_int8_aidlite/python/test.jpg ADDED
model_farm_facial_landmark_det_qcs8550_qnn2.31_w8a16_aidlite/README.md ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## Model Information
2
+
3
+ ### Source model
4
+ - Input shape: [1x3x128x128]
5
+ - Number of parameters: 5.17M
6
+ - Model size: 20.95M
7
+ - Output shape: [1x265]
8
+
9
+ Source model repository: [Facial-Landmark-Detection](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/facemap_3dmm/model.py)
10
+
11
+ ### Converted model
12
+
13
+ - Precision: W8A16
14
+ - Backend: QNN2.31
15
+ - Target Device: SNM972 QCS8550
16
+
17
+ ## Inference with AidLite SDK
18
+
19
+ ### SDK installation
20
+ Model Farm uses AidLite SDK as the model inference SDK. For details, please refer to the [AidLite Developer Documentation](https://v2.docs.aidlux.com/en/sdk-api/aidlite-sdk/)
21
+
22
+ - Install AidLite SDK
23
+
24
+ ```bash
25
+ # Install the appropriate version of the aidlite sdk
26
+ sudo aid-pkg update
27
+ sudo aid-pkg install aidlite-sdk
28
+ # Download the qnn version that matches the above backend. Eg Install QNN2.23 Aidlite: sudo aid-pkg install aidlite-qnn223
29
+ sudo aid-pkg install aidlite-{QNN VERSION}
30
+ ```
31
+
32
+ - Verify AidLite SDK
33
+
34
+ ```bash
35
+ # aidlite sdk c++ check
36
+ python3 -c "import aidlite ; print(aidlite.get_library_version())"
37
+
38
+ # aidlite sdk python check
39
+ python3 -c "import aidlite ; print(aidlite.get_py_library_version())"
40
+ ```
model_farm_facial_landmark_det_qcs8550_qnn2.31_w8a16_aidlite/converted_models/facial_landmark_det_w8a16.qnn231.ctx.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8858ab712b6a31f93cfad48ac7d41c0080100b4be47d82f79ea61c418dc98055
3
+ size 5552672
model_farm_facial_landmark_det_qcs8550_qnn2.31_w8a16_aidlite/python/test.jpg ADDED