Invalid protobuf

[sametucuncu]

Hello! I am getting this error. It would be appreciated if you could help with the problem.

onnxruntime.capi.onnxruntime_pybind11_state.InvalidProtobuf: [ONNXRuntimeError]: 7: INVALID_PROTOBUF: Load model from assets/face_detector.onnx failed:Protobuf parsing failed.

[adamswsk]

我也遇到了相同的问题，我用的是python 3.8.10，按照文档安装了依赖，运行main.py的时候报错了，是我的环境有问题么？

[yinguobing]

Please check the model file with md5sum

md5sum face_detector.onnx

The output should be

9d44dbc74e862957c4cf3b81294b9c94  face_detector.onnx

If the digest value differs from above, the model file was corrupted. Try clone again, or download directly from github.

[sametucuncu]

Thank you for your help. You mean download the pretrained model from SCRFD and convert it to onnx file and replace it with the corrupted one.

[adamswsk]

Thank you for your response. I have verified that the MD5 hash value is correct. Based on sametucuncu's suggestion, I am considering using the pretrained model from SCRFD and converting it to an ONNX file

[sametucuncu]

Thank you for your response. I have verified that the MD5 hash value is correct. Based on sametucuncu's suggestion, I am considering using the pretrained model from SCRFD and converting it to an ONNX file

Could you please let me know if you could convert the model to onnx file?

[AnthonyLia]

I am having this same issue, Windows with Python 3.11 environment.
The first time, face_detector.onnx had an MD5 hash of a117d2bade61178baca5b6fa3163e72a
After redownloading, it's 9d44dbc74e862957c4cf3b81294b9c94

However, I still receive:

 Traceback (most recent call last):
  File "c:\Users\A\Desktop\head-pose-estimation-master\main.py", line 123, in <module>
    run()
  File "c:\Users\A\Desktop\head-pose-estimation-master\main.py", line 49, in run
    mark_detector = MarkDetector("assets/face_landmarks.onnx")
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "c:\Users\A\Desktop\head-pose-estimation-master\mark_detection.py", line 20, in __init__
    self.model = ort.InferenceSession(
                 ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\A\AppData\Roaming\Python\Python311\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 383, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "C:\Users\A\AppData\Roaming\Python\Python311\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 424, in _create_inference_session
    sess = C.InferenceSession(session_options, self._model_path, True, self._read_config_from_model)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
onnxruntime.capi.onnxruntime_pybind11_state.InvalidProtobuf: [ONNXRuntimeError] : 7 : INVALID_PROTOBUF : Load model from assets/face_landmarks.onnx failed:Protobuf parsing failed.

as the traceback, but the 2nd run my console printed:

2 0 2 4 - 0 2 - 0 4   2 1 : 5 5 : 2 5 . 2 5 8 6 1 7 9   [ E : o n n x r u n t i m e : D e f a u l t ,   p r o v i d e r _ b r i d g e _ o r t . c c : 1 3 5 1   o n n x r u n t i m e : : T r y G e t P r o v i d e r I n f o _ C U D A ]   D : \ a \ _ w o r k \ 1 \ s \ o n n x r u n t i m e \ c o r e \ s e s s i o n \ p r o v i d e r _ b r i d g e _ o r t . c c : 1 1 3 1   o n n x r u n t i m e : : P r o v i d e r L i b r a r y : : G e t   [ O N N X R u n t i m e E r r o r ]   :   1   :   F A I L   :   L o a d L i b r a r y   f a i l e d   w i t h   e r r o r   1 2 6   " "   w h e n   t r y i n g   t o   l o a d   " C : \ U s e r s \ A \ A p p D a t a \ R o a m i n g \ P y t h o n \ P y t h o n 3 1 1 \ s i t e - p a c k a g e s \ o n n x r u n t i m e \ c a p i \ o n n x r u n t i m e _ p r o v i d e r s _ c u d a . d l l "

 2 0 2 4 - 0 2 - 0 4   2 1 : 5 5 : 2 5 . 2 7 8 9 9 0 1   [ W : o n n x r u n t i m e : D e f a u l t ,   o n n x r u n t i m e _ p y b i n d _ s t a t e . c c : 6 4 0   o n n x r u n t i m e : : p y t h o n : : C r e a t e E x e c u t i o n P r o v i d e r I n s t a n c e ]   F a i l e d   t o   c r e a t e   C U D A E x e c u t i o n P r o v i d e r .   P l e a s e   r e f e r e n c e   h t t p s : / / o n n x r u n t i m e . a i / d o c s / e x e c u t i o n - p r o v i d e r s / C U D A - E x e c u t i o n P r o v i d e r . h t m l # r e q u i r e m e n t s   t o   e n s u r e   a l l   d e p e n d e n c i e s   a r e   m e t .

Is cuda required to run this?

EDIT:
After realizing the script went to mark_detector = MarkDetector("assets/face_landmarks.onnx") the 2nd time instead of failing at face_detector = FaceDetector("assets/face_detector.onnx") I redownloaded face_landmarks.onnx as well and now it works, albeit with the cuda console errors.

[IDEALIST-MAKER]

I am having this same issue, Windows with Python 3.11 environment. The first time, face_detector.onnx had an MD5 hash of a117d2bade61178baca5b6fa3163e72a After redownloading, it's 9d44dbc74e862957c4cf3b81294b9c94

However, I still receive:

 Traceback (most recent call last):
  File "c:\Users\A\Desktop\head-pose-estimation-master\main.py", line 123, in <module>
    run()
  File "c:\Users\A\Desktop\head-pose-estimation-master\main.py", line 49, in run
    mark_detector = MarkDetector("assets/face_landmarks.onnx")
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "c:\Users\A\Desktop\head-pose-estimation-master\mark_detection.py", line 20, in __init__
    self.model = ort.InferenceSession(
                 ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\A\AppData\Roaming\Python\Python311\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 383, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "C:\Users\A\AppData\Roaming\Python\Python311\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 424, in _create_inference_session
    sess = C.InferenceSession(session_options, self._model_path, True, self._read_config_from_model)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
onnxruntime.capi.onnxruntime_pybind11_state.InvalidProtobuf: [ONNXRuntimeError] : 7 : INVALID_PROTOBUF : Load model from assets/face_landmarks.onnx failed:Protobuf parsing failed.

as the traceback, but the 2nd run my console printed:

2 0 2 4 - 0 2 - 0 4   2 1 : 5 5 : 2 5 . 2 5 8 6 1 7 9   [ E : o n n x r u n t i m e : D e f a u l t ,   p r o v i d e r _ b r i d g e _ o r t . c c : 1 3 5 1   o n n x r u n t i m e : : T r y G e t P r o v i d e r I n f o _ C U D A ]   D : \ a \ _ w o r k \ 1 \ s \ o n n x r u n t i m e \ c o r e \ s e s s i o n \ p r o v i d e r _ b r i d g e _ o r t . c c : 1 1 3 1   o n n x r u n t i m e : : P r o v i d e r L i b r a r y : : G e t   [ O N N X R u n t i m e E r r o r ]   :   1   :   F A I L   :   L o a d L i b r a r y   f a i l e d   w i t h   e r r o r   1 2 6   " "   w h e n   t r y i n g   t o   l o a d   " C : \ U s e r s \ A \ A p p D a t a \ R o a m i n g \ P y t h o n \ P y t h o n 3 1 1 \ s i t e - p a c k a g e s \ o n n x r u n t i m e \ c a p i \ o n n x r u n t i m e _ p r o v i d e r s _ c u d a . d l l "

 2 0 2 4 - 0 2 - 0 4   2 1 : 5 5 : 2 5 . 2 7 8 9 9 0 1   [ W : o n n x r u n t i m e : D e f a u l t ,   o n n x r u n t i m e _ p y b i n d _ s t a t e . c c : 6 4 0   o n n x r u n t i m e : : p y t h o n : : C r e a t e E x e c u t i o n P r o v i d e r I n s t a n c e ]   F a i l e d   t o   c r e a t e   C U D A E x e c u t i o n P r o v i d e r .   P l e a s e   r e f e r e n c e   h t t p s : / / o n n x r u n t i m e . a i / d o c s / e x e c u t i o n - p r o v i d e r s / C U D A - E x e c u t i o n P r o v i d e r . h t m l # r e q u i r e m e n t s   t o   e n s u r e   a l l   d e p e n d e n c i e s   a r e   m e t .

Is cuda required to run this?

EDIT: After realizing the script went to mark_detector = MarkDetector("assets/face_landmarks.onnx") the 2nd time instead of failing at face_detector = FaceDetector("assets/face_detector.onnx") I redownloaded face_landmarks.onnx as well and now it works, albeit with the cuda console errors.

Could you tell me where I can download the face_detector.onnx and face_loadmarks.onnx? I use command "git lfs clone",but it doesn't make it.

[yinguobing]

The onnx files are tracked by git lfs. Please fetch them with git lfs after cloning the repository

git lfs pull

If you see error message like this

batch response: This repository is over its data quota. Account responsible for LFS bandwidth should purchase more data packs to restore access.                                         
error: failed to fetch some objects from 'https://github.com/yinguobing/head-pose-estimation.git/info/lfs'

Unfortunately this is due to Github data usage quota, which provides 1 GB per month/repo. And yes, I'm also blocked by this quota limitation.

It seems we all have to wait for the next month's quota reset, then I can upload the onnx files to the release page that is free from the usage quota.

I'm sorry for this inconvenience.

[Sadat75]

Dear @yinguobing,

Thank you for your work. Currently, both the detection and the landmark onnx files seem to be corrupted. Is it possible to upload the onnx file for the facial landmark model in a drive this month?

Thank you.

[yinguobing]

I'm afraid that's not possible. There is no local backup of this repo since I used to believe GitHub is a safer place. Currently we all have to wait.

[Sadat75]

Dear @yinguobing,

Thank you for the response. I am looking forward to the availability of the facial landmark model soon.

[yinguobing]

Models are available in release page:
https://github.com/yinguobing/head-pose-estimation/releases/tag/v1.0.0