opencv-contrib-python 4.0.0 can not read caffe model and prototxt

[jozhuzhu]

System information (version)

• OpenCV => :4.0.0.21:
• Operating System / Platform => : windows10:
• Compiler => :pre-build version from pip:

Detailed description

background: i wanna achieve opencv dnn model to detect face in an image
after accessing the model and prototxt from opencv_extra/testdata/dnn/download_models.py, i learned some sample code to test whether it works or not. And then, I got the following info:

[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 2:1: Invalid control characters encountered in text.
[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 2:2: Interpreting non ascii codepoint 162.
[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 2:2: Expected identifier, got: ?
Traceback (most recent call last):
  File "src/facedetect/detector.py", line 103, in <module>
    util.gainFaceByDNN(path, modelpath, deploypath)
  File "src/facedetect/detector.py", line 26, in gainFaceByDNN
    net = cv2.dnn.readNetFromCaffe(modelFile, configFile)
cv2.error: OpenCV(4.0.0) C:\projects\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1151: error: (-2:Unspecified error) FAILED: ReadProtoFromTextFile(param_file, param). Failed to parse NetParameter file: D:\project\python\IQA\src\facedetect\res10_300x300_ssd_iter_140000.caffemodel in function 'cv::dnn::ReadNetParamsFromTextFileOrDie'

i have tried to google error, and got some solutions to try but they all failed, including checking the invisible content of the deploy.prototxt, redownload the files, etc.

Steps to reproduce

below is the code i used for test and it goes wrong on the first code.

net = cv2.dnn.readNetFromCaffe(modelFile, configFile)
image = cv2.imread(imgPath)
(h, w) = image.shape[:2]

blob = cv2.dnn.blobFromImage(cv2.resize(image, (300,300)), 1.0, (300,300), (103.93, 116.77, 123.68), False)
net.setInput(blob)

detections = net.forward()

Is there any practical solutions ? thanks ~

[dkurt]

@jozhuzhu, arguments are in incorrect order:

net = cv2.dnn.readNetFromCaffe(modelFile, configFile)

see https://docs.opencv.org/master/d6/d0f/group__dnn.html#ga29d0ea5e52b1d1a6c2681e3f7d68473a

Change it to

net = cv2.dnn.readNetFromCaffe(configFile, modelFile)

or use readNet with any order.

[jozhuzhu]

thanks you very much!!!

[eliaed]

hello i also have a problem with the file pose_iter_160000.caffemodel" is failed.
the code is:
import cv2
import time
import numpy as np
protoFile = "pose/mpi/pose_deploy_linevec_faster_4_stages.prototxt"
weightsFile = "pose/mpi/pose_iter_160000.caffemodel"
nPoints = 15
POSE_PAIRS = [[0, 1], [1, 2], [2, 3], [3, 4], [1, 5], [5, 6], [6, 7], [1, 14], [14, 8], [8, 9], [9, 10], [14, 11],
[11, 12], [12, 13]]

frame = cv2.imread("single.jpeg")
frameCopy = np.copy(frame)
frameWidth = frame.shape[1]
frameHeight = frame.shape[0]
threshold = 0.1

net = cv2.dnn.readNetFromCaffe(protoFile, weightsFile)

the failed message is:
[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 7:1: Expected identifier, got: <
Traceback (most recent call last):
File "C:/Users/elia/PycharmProjects/Image/test.py", line 16, in
net = cv2.dnn.readNetFromCaffe(protoFile, weightsFile)
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1151: error: (-2:Unspecified error) FAILED: ReadProtoFromTextFile(param_file, param). Failed to parse NetParameter file: pose/mpi/pose_deploy_linevec_faster_4_stages.prototxt in function 'cv::dnn::ReadNetParamsFromTextFileOrDie'

thank you!!!!!!!!!!

[noorkhokhar99]

System information (version)

• OpenCV => :4.0.0.21:
• Operating System / Platform => : windows10:
• Compiler => :pre-build version from pip:

Detailed description

background: i wanna achieve opencv dnn model to detect face in an image
after accessing the model and prototxt from opencv_extra/testdata/dnn/download_models.py, i learned some sample code to test whether it works or not. And then, I got the following info:

[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 2:1: Invalid control characters encountered in text.
[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 2:2: Interpreting non ascii codepoint 162.
[libprotobuf ERROR C:\projects\opencv-python\opencv\3rdparty\protobuf\src\google\protobuf\text_format.cc:288] Error parsing text-format opencv_caffe.NetParameter: 2:2: Expected identifier, got: ?
Traceback (most recent call last):
  File "src/facedetect/detector.py", line 103, in <module>
    util.gainFaceByDNN(path, modelpath, deploypath)
  File "src/facedetect/detector.py", line 26, in gainFaceByDNN
    net = cv2.dnn.readNetFromCaffe(modelFile, configFile)
cv2.error: OpenCV(4.0.0) C:\projects\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1151: error: (-2:Unspecified error) FAILED: ReadProtoFromTextFile(param_file, param). Failed to parse NetParameter file: D:\project\python\IQA\src\facedetect\res10_300x300_ssd_iter_140000.caffemodel in function 'cv::dnn::ReadNetParamsFromTextFileOrDie'

i have tried to google error, and got some solutions to try but they all failed, including checking the invisible content of the deploy.prototxt, redownload the files, etc.

Steps to reproduce

below is the code i used for test and it goes wrong on the first code.

net = cv2.dnn.readNetFromCaffe(modelFile, configFile)
image = cv2.imread(imgPath)
(h, w) = image.shape[:2]

blob = cv2.dnn.blobFromImage(cv2.resize(image, (300,300)), 1.0, (300,300), (103.93, 116.77, 123.68), False)
net.setInput(blob)

detections = net.forward()

Is there any practical solutions ? thanks ~

you can try it.
pip install opencv-python==3.1.0.0

enjoy it!.

---

Update by @alalek:

pip install opencv-python==3.1.0.0

OpenCV 3.1.0 doesn't have DNN module.

[Soumi7]

i get this error and it isnt resolved

[berak]

@Suomi7, please do not append to outdated and closed issues.

instead, ask on https://forum.opencv.org/, where we can help you much better
(given, you supply us opencv version, your code, a link to the model, the whole chi-chi...)
((please also make sure, you have valid files, not some html error msgs, like the error suggests))

[mfahmifadh]

i've the same problem

my code

rototxtPath = r"face_detector\deploy.prototxt"
weightsPath = r"face_detector\res10_300x300_ssd_iter_140000.caffemodel"
faceNet = cv2.dnn.readNet(prototxtPath, weightsPath)

maskNet = load_model("mask_detector.model")

and i got error

---------------------------------------------------------------------------
error                                     Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_13672/2145281415.py in <module>
     34 prototxtPath = r"face_detector\deploy.prototxt"
     35 weightsPath = r"face_detector\res10_300x300_ssd_iter_140000.caffemodel"
---> 36 faceNet = cv2.dnn.readNet(prototxtPath, weightsPath)
     37 
     38 maskNet = load_model("mask_detector.model")

error: OpenCV(4.5.4) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1126: error: (-2:Unspecified error) FAILED: fs.is_open(). Can't open "face_detector\deploy.prototxt" in function 'cv::dnn::ReadProtoFromTextFile'

i tried searching on google with the same problem, and i got problem in certain file. My python project file is in C:\Users\mfahm\anaconda3\Test. Am I getting the wrong filename?

[alalek]

fs.is_open(). Can't open

Wrong path is specified. File is missing or no access rights.
(nothing to fix in OpenCV)

---

Usage questions should go to Users OpenCV Q/A forum: https://forum.opencv.org/
This tracker is for issues and bugs that needs fix in OpenCV.

[eliaed]

thank you !! ‫בתאריך יום ו׳, 26 בנוב׳ 2021 ב-7:34 מאת ‪Alexander Alekhin‬‏ <‪ ***@***.***‬‏>:‬

…

fs.is_open(). Can't open Wrong path is specified. File is missing or no access rights. (nothing to fix in OpenCV) ------------------------------ Usage questions should go to Users OpenCV Q/A forum: https://forum.opencv.org/ This tracker is for issues and bugs that needs fix in OpenCV. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#13961 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKZKOW4JUUCADCLGBJDTESTUN4L7RANCNFSM4G3OX5VA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.

[njwy1978]

I doenload "ResNet50_deploy.caffemodel" and "ResNet50_deploy.prototxt", then use the following code:
net = cv2.dnn.readNetFromCaffe(ResNet50_deploy.prototxt, ResNet50_deploy.caffemodel)
I got the eror:
FAILED: ReadProtoFromBinaryFile(param_file, param). Failed to parse NetParameter file: ./mobilenet_ssd/ResNet50_deploy.caffemodel in function 'cv::dnn::ReadNetParamsFromBinaryFileOrDie'
how can I solve this problem?

[feersdilaa]

hello sir, excuse me, can you help me?
here my code :

import numpy as np
import cv2

prototxt_path = 'py/colorization_deploy_v2.prototxt'
model_path = 'py/colorization_release_v2.caffemodel'
kernel_path = 'py/pts_in_hull.npy'
image_path = 'tester.jpg'

net = cv2.dnn.readNetFromCaffe(prototxt_path, model_path)
points = np.load(kernel_path)

points = points.transpose().reshape(2, 313, 1, 1)
net.getLayer(net.getLayerId("class_ab")).blobs = [points.astype(np.float)]
net.getLayer(net.getLayerId("conv8_313_rh")).blobs = [np.full([1, 313], 2.206, dtype="float32")]

bw_image = cv2.imread(image_path)
normalized = bw.image.astype("float32") / 255.0
lab = cv2.cvtCololr(normalized, cv2.COLOR_BGR2LAB)

resized = cv2.resize(lab, (224, 224))

L = cv2.split(resized)[0]
L -= 50

net.setInput(cv2.dnn.blobFromImage(L))
ab = net.forward()[0, :, :, :].transpose((1,2,0))

ab = cv2.resize(ab, (bw_image.shape[1], bw_image[0]))
L = cv2.split(lab)[0]

colorized = np.concatenate((L[:,:,np.newaxis], ab), axis=2)
colorized = cv2.cvtColor(colorized, cv.COLOR_BGR2LAB)
colorized = (255.0 * colorized).astype("uint8")

cv2.imshow("BW Image", bw_image)
cv2.imshow("Colorized", colorized)
cv2.waitKey(0)
cv2.destroyAllWindows()

and im got error :

net = cv2.dnn.readNetFromCaffe(prototxt_path, model_path)
cv2.error: OpenCV(4.7.0) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1126: error: (-2:Unspecified error)
FAILED: fs.is_open(). Can't open "py/colorization_deploy_v2.prototxt" in function 'cv::dnn::ReadProtoFromTextFile'

thankyou sir!

[ibtehajali67]

My problem is solved just updating the right path

protoFile = "pose\mpi\pose_deploy_linevec_faster_4_stages.prototxt"
weightsFile = "pose\mpi\pose_iter_160000.caffemodel"
net = cv2.dnn.readNetFromCaffe(protoFile, weightsFile)

[meaadMa]

hello sir, excuse me, can you help me? here my code :

import numpy as np import cv2

prototxt_path = 'py/colorization_deploy_v2.prototxt' model_path = 'py/colorization_release_v2.caffemodel' kernel_path = 'py/pts_in_hull.npy' image_path = 'tester.jpg'

net = cv2.dnn.readNetFromCaffe(prototxt_path, model_path) points = np.load(kernel_path)

points = points.transpose().reshape(2, 313, 1, 1) net.getLayer(net.getLayerId("class_ab")).blobs = [points.astype(np.float)] net.getLayer(net.getLayerId("conv8_313_rh")).blobs = [np.full([1, 313], 2.206, dtype="float32")]

bw_image = cv2.imread(image_path) normalized = bw.image.astype("float32") / 255.0 lab = cv2.cvtCololr(normalized, cv2.COLOR_BGR2LAB)

resized = cv2.resize(lab, (224, 224))

L = cv2.split(resized)[0] L -= 50

net.setInput(cv2.dnn.blobFromImage(L)) ab = net.forward()[0, :, :, :].transpose((1,2,0))

ab = cv2.resize(ab, (bw_image.shape[1], bw_image[0])) L = cv2.split(lab)[0]

colorized = np.concatenate((L[:,:,np.newaxis], ab), axis=2) colorized = cv2.cvtColor(colorized, cv.COLOR_BGR2LAB) colorized = (255.0 * colorized).astype("uint8")

cv2.imshow("BW Image", bw_image) cv2.imshow("Colorized", colorized) cv2.waitKey(0) cv2.destroyAllWindows()

and im got error :

net = cv2.dnn.readNetFromCaffe(prototxt_path, model_path) cv2.error: OpenCV(4.7.0) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1126: error: (-2:Unspecified error) FAILED: fs.is_open(). Can't open "py/colorization_deploy_v2.prototxt" in function 'cv::dnn::ReadProtoFromTextFile'

thankyou sir!

I faced the same problem previously and the solution was just to write the full path as:
PROTOTXT = 'C:\Users\LEGION\Documents\Meaad\color\colorization_deploy_v2.prototxt'
POINTS = 'C:\Users\LEGION\Documents\Meaad\color\pts_in_hull.npy'