OpenCV Yolo V3 tiny - opencv

I am trying to use Yolo tiny on Open CV 3.4.1 Can anyone confirm whether it is supported?
I get an error when reading the config file - I am using yolov3-tiny.weights and yolov3-tiny.cfg downloaded from
https://github.com/pjreddie/darknet/blob/master/cfg/yolov3-tiny.cfg
E/cv::error(): OpenCV(3.4.1) Error: Parsing error (Unknown layer type: yolo) in bool cv::dnn::darknet::ReadDarknetFromCfgFile(const char*, cv::dnn::darknet::NetParameter*), file /build/master_pack-android/opencv/modules/dnn/src/darknet/darknet_io.cpp, line 503 07-02 12:10:21.455 31240-31240/ E/org.opencv.dnn: dnn::readNetFromDarknet_10() caught cv::Exception: OpenCV(3.4.1) /build/master_pack-android/opencv/modules/dnn/src/darknet/darknet_io.cpp:503: error: (-212) Unknown layer type: yolo in function bool cv::dnn::darknet::ReadDarknetFromCfgFile(const char*, cv::dnn::darknet::NetParameter*)

As of April 16, yes, it is supported (pull request here). I don't know what your code looks like, but it seems like someone else had the same problem and was able to resolve it. If this doesn't help, feel free to post some code you have and we can give it a look.

Related

Ruby gem origami sign signature failed

I'm trying to make the PDF file with signature with origami gem, follow this example https://github.com/gdelugre/origami/blob/master/examples/signature/signature.rb
Now i just run this signature.rb and got error
[error] Breaking on: ">>\nendobj\n..." at offset 0x1f6f3
[error] Last exception: [Origami::InvalidObjectError] Failed to parse object (no:43,gen:0) -> [ArgumentError] wrong number of arguments (given 1, expected 0; required keyword: year)
I have no idea to move forward :'<
I also found the sign method of gem at https://github.com/gdelugre/origami/blob/master/lib/origami/signature.rb, and have take a look, I can't find any specific things to do :'<
Might this example is outdate?
The error message is known see https://github.com/gdelugre/origami/issues/80
A fix should be available https://github.com/gdelugre/origami/pull/74/commits
But has not been added so use newer file from the fork https://github.com/pocke/origami/tree/fix-ruby-2.7-kwargs-warnings
Specifically you need this updated file and may need to follow any other suggestions from above.
https://github.com/pocke/origami/raw/fix-ruby-2.7-kwargs-warnings/lib/origami/filters/predictors.rb
However it always worth looking for a fork with many more recent improvements such as
https://github.com/joelsondrew/origami

TensorflowJS TFJS error: The dtype of dict

Had tried to run https://glitch.com/~tar-understood-exoplanet
and the model would fail to load and I wouldn't be able to use enable the webcam.
Anyone had the same issue?
While the program is running, in the console I get the following:
tfjs:2 Uncaught (in promise) Error: The dtype of dict['image_tensor'] provided in model.execute(dict) must be int32, but was float32
at Object.b [as assert] (tfjs:2)
at tfjs:2
at Array.forEach (<anonymous>)
at t.checkInputShapeAndType (tfjs:2)
at t.<anonymous> (tfjs:2)
at tfjs:2
at Object.next (tfjs:2)
at tfjs:2
at new Promise (<anonymous>)
at Zv (tfjs:2)
I have a Macbook Pro and some other people on their Windows also had some issues running the model. We also tried it on different browsers, Safari and Chrome.
SUCCESS! after switching to coco-ssd 2.0.2:
I added the version 2.0.2 in line 62 as follows:
<script src="https://cdn.jsdelivr.net/npm/#tensorflow-models/coco-ssd#2.0.2"></script>
This is caused by the warmup run of coco-ssd that uses tf.zeros tensor. The default dtype for tf.zeros is 'float' in the recent release of TFJS.
I have put out a new version with fixes. It should work if you use the latest version of coco-ssd (2.0.2) in the glitch example (index.html) as following.
<!-- Load the coco-ssd model to use to recognize things in images -->
<script src="https://cdn.jsdelivr.net/npm/#tensorflow-models/coco-ssd#2.0.2"></script>
Same error here, just occured since Friday night (04/03/2020)
TFModel works well in past few weeks.
I got the same error.
My Scenerio:
I trained a pre-trained model from tensorflow model zoo using transfer learning using tensorflow api as saved model (model.pb file) and converted it into tfjs format (model.json and shared .bin files).
When I tried running this model.json on the javascript(web), it gives the below error:
Uncaught (in promise) Error: The dtype of dict['input_tensor'] provided in model.execute(dict) must be int32, but was float32
When I tried someone else's working converted model (model.json and shared .bin files) on my javascript(web), it worked.
Conclusion:
There is something wrong with my converted model. I converted it using tensorflowjs_converter. My original model in (model.pb) works accurately in python too.
I'm still trying out to convert my model.pb file with different tensorflowjs_converters as it seems to be the converters versioning issue.

How to setup jpeglib with MS Visual C++ 2010 Express

I try to use the well known "jpeglib" now available in version 9.
Overview: I am a programming starter in C using MS Visual C++ 2010 Express and read a C-book finally. My goal is to compress a RGB(.bmp)-file into JPEG. So I downloaded the jpeglib (v9), read all the install and so on *.txt-files. It was possible to compile the C-files to get cjpeg.exe, djpeg.exe, jpeg.lib and so on.
After this success I tried to use the example.c to get deeper into the jpeg and to control the compression. -> It did not work. Then I tried to use the jpeglib included in OpenCV. Same non working result, while the OpenCV code samples are compiled successfully.
I googled lots of samples, problem solutions and tips how to use jpeglib. Most helpful information I found here. But in the result I was still not able to compile my own jpeg-tool or example.c (rewriting the write_JPEG_file() function ).
Question: What in the basics could be wrong, when every other programmer just uses #include "jpeglib.h" in their codes to run the jpeg functions successfully? For me including this file seems not to be enough. Because of that I have to ask you and to relaunch a one year old Question that also did not helped solving my probs. I can not give you a specific error message because of multiple attempts to solve those issues.
Here is my current MS-VC++2010 config for using OpenCV 2.4.6:
C/C++ -> General -> Additional Include Directories:
C:\opencv\build\include
C:\opencv\build\include\opencv
C:\opencv\build\include\opencv2
C:\opencv\include
C:\opencv\include\opencv
C:\opencv\include\opencv2
C:\opencv\3rdparty\libjpeg
Linker -> General -> Additional Library Directories:
C:\opencv\build\x86\vc10\lib
Linker -> Input -> Additional Dependencies:
opencv_core246d.lib
opencv_imgproc246d.lib
opencv_highgui246d.lib
opencv_ml246d.lib
opencv_video246d.lib
opencv_features2d246d.lib
opencv_calib3d246d.lib
opencv_objdetect246d.lib
opencv_contrib246d.lib
opencv_legacy246d.lib
opencv_flann246d.lib
jpeg.lib [this I added into C:\opencv\build\x86\vc10\lib]
At last some threads I have already read:
Compressing IplImage to JPEG using libjpeg in OpenCV - 1
Compressing IplImage to JPEG using libjpeg in OpenCV - 2
Need help in reading JPEG file using libjpeg
how can i use jpeg_mem_src, jpeg_mem_dest in libjpeg-turbo?
If there is something missing you want to know I will do my best to add this information!
EDIT - Some error messages:
1>jpeg_coder.obj : error LNK2001: unresolved external symbol ""unsigned char * image_buffer" (?image_buffer##3PAEA)".
1>jpeg_coder.obj : error LNK2001: unresolved external symbol ""int image_height" (?image_height##3HA)".
1>jpeg_coder.obj : error LNK2001: unresolved external symbol """int image_width" (?image_width##3HA)".
First symbol should be typedefed in jpeglib.h as JSAMPLE FAR *JSAMPROW,
second and third are defined in rdjpgcom.c as unsigned int image_height, image_width;
Maybe, with a lot of luck, I figured out what is need to be done to get jpeglib running. In the current state I have to check and double check whether my problem solution is correct or not. At this point thank you very much for your attention for my issues. As soon as I am sure with the solution I will post it.
Problem Solution: I did not setup OpenCV correctly for using jpeglib. The mistake I made was to link the wrong *.lib-file. The proper file for linking is libjpegd.lib. Too bad that jpeglib v6.2 is included in OpenCV 2.4.6 instead of v9. The hint to add HAVE_JPEG to use jpeglib functions inside OpenCV I ignored completely. For now the problem is solved.
Problem Solution: I did not setup OpenCV correctly for using jpeglib. The mistake I made was to link the wrong *.lib-file.
The proper file for linking is libjpegd.lib.
Too bad that jpeglib v6.2 is included in OpenCV 2.4.6 instead of v9. The hint to add HAVE_JPEG to use jpeglib functions inside OpenCV I ignored completely. Thanks to all who tried solving this issue. For now the problem is solved.

Cannot get SURF example in EMGU.CV to work?

I am trying to detect a pattern shown in two images. Hence I have been trying to use the SURF algorithim found in emgu.CV, but the "SURFFeature" example that is given gives me the following error:
An unhandled exception of type 'Emgu.CV.Util.CvException' occurred in Emgu.CV.dll
Additional information: OpenCV: norm == NORM_L1 || norm == NORM_L2 || norm == NORM_HAMMING
Any ideas how to fix this?
When I try the "Hello World" example and the face detection example, both seem to work fine.
Thanks for any advice!
Fouad.
PS: Emgu.CV can be downloaded from here: http://www.emgu.com/wiki/index.php/Main_Page
Apparently the build was messed up.
http://www.emgu.com/bugs/show_bug.cgi?format=multiple&id=74
Aha, found it. The error here is in Emgu.Cv.Gpu/GpuBruteForceMatcher.cs lines 22 and 27.
Line 22 currently reads:
L2Dist,
It should read: L2Dist = 4,
Line 27 currently reads: HammingDist
It should read: HammingDist = 6
Rebuild the Emgu.CV.Gpu dll with those changes and it works.

Simple OpenCV problem

Why I try to run the following OpenCV program, it shows the following error :
ERROR:
test_1.exe - Application Error
The application failed to initialize properly (0x80000003).
Click on OK to terminate the application.
CODE:
#include "cv.h"
#include "highgui.h"
int main()
{
IplImage *img = cvLoadImage("C:\\face.bmp");
cvSetImageROI(img, cvRect(100,100, 100, 100));
cvAddS(img, cvScalar(50), img);
cvResetImageROI(img);
cvShowImage("Test", img);
cvWaitKey(0);
return 0;
}
When i press F5(im using vs2008express), the program encounters a break point...i have attached a picture...dont know, whether, it will help or not.
Error Snapshot Link
It is not that, only this program is producing this error, but also any kind of image manipulation funciton containing (OpenCV)program is resulting in this sitution.
Such as : cvSmooth
one last thing, it there any dedicated OpenCV forum or sth like that?
I am an administrator.So, yes, ive the permission.
a version mismatch.
sorry, i didn't get it?Version mismatch with what?
But, i have found the error using dependency walker.
Warning: At least one module has an unresolved import due to a missing export
function in a delay-load dependent module.
and also found that, it is a common problem, and found some info in the FAQ of DW...
Why am I seeing a lot of applications where MPR.DLL shows up in red under
SHLWAPI.DLL because it is missing a function named WNetRestoreConnectionA?
I also get a "Warning: At least one module has an unresolved import due to
a missing export function in a delay-load dependent module" message.
Function name : WNetRestoreConnectionA
But there is no guideline about how to solve it. Though, they say, it is not a problem.
i googled a little and found a suggestion.It says,
Turn off your compilers setting to assume you are programming for Win9x.
(I just lost which setting but it is not that difficult, use a #define...)
But i have no idea, how to do that in Visual Studio 2008 express.
Any suggestion how to solve it...
This usually indicates a problem with a dll; either you don't have permission, or a version is mismatched. Try running as Administrator to see if it is a permissions problem. If that doesn't help, try using the Dependency Walker.

Resources