[RELEASED] OpenCV ObjectDetector

OpenCV ObjectDetector
http://u3d.as/content/enox-software/open-cv-object-detector

Requires Unity 4.5.5 or higher.

[NEW] Support for Unity5

Works with Unity Free & Pro
iOS & Android support
Win & Mac Standalone support
Support for preview in the Editor

More advanced OpenCV Asset is published.Please check OpenCV for Unity.


OpenCV ObjectDetector can detect(Sync or Async) an object from Texture2D using OpenCV.
SampleCode | Demo Application(FaceDetection) | Tutorial & Demo Video(Unity4 Unity5) | Forum

e-mail
enox.software@gmail.com



Features:

• You can get a processing result of detectMultiScale() of OpenCV using cascade file that you specified.
• Object detection parameters (same as the parameters of detectMultiScale()) can be set in JSON format, You can get in JSON format Object detection result.

Overview of the method call steps:

1. LoadCascade (string filename)
2. AddObjectDetectorParam (string param)
3. Detect (Texture2D texture, string callbackGameObjectName, string callbackMethodName)

System Requirements:

Build Win Standalone & Preview Editor : Windows7 or later
Build Mac Standalone & Preview Editor : OSX 10.8 or later​

Release Notes:
1.1.9

[iOS]Fixed libopencvobjectdetector.a Bitcode Setting. ​

1.1.8

[iOS]Enabled Bitcode. ​

1.1.7

[Common] Fixed the bug that occurs in the editor. ​

1.1.6

[iOS]Move “OpenCVObjectDetector/ iOSforXcode/opencv2.framework”to “OpenCVObjectDetector/Plugins/iOS/”folder. ​

1.1.5

[Common]Add SampleScene Setup Tutorial Video for Unity5.​

1.1.4

[Common]Add OpenCVObjectDetectorMenuItem.cs.( This script set plugin import settings automatically from MenuItem.)
[iOS]Move “OpenCVObjectDetector/ iOSforXcode/iOS_BuildPostprocessor.cs” to “OpenCVObjectDetector/Editor”folder. ​

1.1.3

[Common]Update to OpenCV2.4.11​

1.1.2

[Common]Divide asset for Unity4 and Unity5. ​

1.1.1

[Common]Support for Unity5. ​

1.1.0

[Common]Update to OpenCV2.4.10​

1.0.9

[iOS]Support for arm64 build target.(Unity 4.6.1p3 or higher) ​

1.0.8

[Android]Support for x86 build target.(Unity 4.6 or higher)​

1.0.7

[Common]Update SampleScene(Process of converting results of object detection to the 3D position).​

1.0.6

[Common]Support for preview in the Editor.(Pro only)
[Common]Support for Win & Mac Standalone.(Pro only)
[Android]Change of location of the cascade file.Changed to use“Aseets/StreamingAssets/” folder.
[iOS] Add the cascade file to Xcode project is no longer required.Changed to use“Aseets/StreamingAssets/” folder.​

 

is this compatible with web player build?

 

This asset is not compatible to the WebPlayer.

http://docs.unity3d.com/Manual/Plugins.html
Note: On the desktop platforms, plugins are a pro-only feature. For security reasons, plugins are not usable with webplayers.

 

ok, thanks for the answer

by the way, is there any way to do face tracking in web player
I'm already stuck on this for weeks

 

i use Unity 4.5.5p1 and XCode 5.1.1 or 6.1.
export xcode project, then add opencv2.framework.
but build failed, have Linker Error.

ld: warning: directory not found for option '-F/Users/me/Fukui/Assets/OpenCVObjectDetector/iOS'

ld: warning: directory not found for option '-Ffor'

ld: warning: directory not found for option '-FXcode'

Undefined symbols for architecture armv7:

"std::__1::__vector_base_common<true>::__throw_length_error() const", referenced from:

std::__1::vector<int, std::__1::allocator<int> >::__append(unsigned long) in opencv2(histogram.o)

std::__1::vector<cv::Vec<int, 128>, std::__1::allocator<cv::Vec<int, 128> > >::__append(unsigned long) in opencv2(matrix.o)

std::__1::vector<cv::Vec<int, 64>, std::__1::allocator<cv::Vec<int, 64> > >::__append(unsigned long) in opencv2(matrix.o)

std::__1::vector<cv::Vec<int, 32>, std::__1::allocator<cv::Vec<int, 32> > >::__append(unsigned long) in opencv2(matrix.o)

std::__1::vector<cv::Vec<int, 16>, std::__1::allocator<cv::Vec<int, 16> > >::__append(unsigned long) in opencv2(matrix.o)

std::__1::vector<cv::Vec<int, 12>, std::__1::allocator<cv::Vec<int, 12> > >::__append(unsigned long) in opencv2(matrix.o)

std::__1::vector<cv::Vec<int, 9>, std::__1::allocator<cv::Vec<int, 9> > >::__append(unsigned long) in opencv2(matrix.o)

...

"cv::_InputArray::_InputArray(cv::Mat const&)", referenced from:

_OpenCVObjectDetector_Detect in libopencvobjectdetector.a(opencvobjectdetector.o)

_DetectThread in libopencvobjectdetector.a(opencvobjectdetector.o)

"cv::_OutputArray::_OutputArray(cv::Mat&)", referenced from:

_OpenCVObjectDetector_Detect in libopencvobjectdetector.a(opencvobjectdetector.o)

_DetectThread in libopencvobjectdetector.a(opencvobjectdetector.o)

"cv::CascadeClassifier::CascadeClassifier(std::string const&)", referenced from:

_OpenCVObjectDetector_LoadCascade in libopencvobjectdetector.a(opencvobjectdetector.o)

ld: symbol(s) not found for architecture armv7

clang: error: linker command failed with exit code 1 (use -v to see invocation)

 

Sorry.
I was able to fix it myself.

 

Hello,
is it possibile to have a mouth+eyes detection on a webtexture playing?

 

This asset can detect the object from only Texture2D.
You can use the ”OpenCV for Unity”, which is a more advanced asset, you can real-time detect the object from WebCamTexture.
https://www.assetstore.unity3d.com/en/#!/content/21088

 

Released Version 1.0.7

Version changes
1.0.7
[Common]Update SampleScene(Process of converting results of object detection to the 3D position).

 

Help me!
i use Unity 4.5.5p1 and XCode 6.1.1 and OS X Yosemte version 10.10.1
when i export xcode project. i have add lib: opencv2.framework.
but build failed, have Linker Error.
Undefined symbols for architecture i386:

"cv::_InputArray::_InputArray(cv::Mat const&)", referenced from:

_OpenCVObjectDetector_Detect in libopencvobjectdetector.a(opencvobjectdetector.o)

_DetectThread in libopencvobjectdetector.a(opencvobjectdetector.o)

"cv::_OutputArray::_OutputArray(cv::Mat&)", referenced from:

_OpenCVObjectDetector_Detect in libopencvobjectdetector.a(opencvobjectdetector.o)

_DetectThread in libopencvobjectdetector.a(opencvobjectdetector.o)

"cv::CascadeClassifier::CascadeClassifier(std::string const&)", referenced from:

_OpenCVObjectDetector_LoadCascade in libopencvobjectdetector.a(opencvobjectdetector.o)

"std::__1::__vector_base_common<true>::__throw_length_error() const", referenced from:

_cvCalcArrHist in opencv2(histogram.o)

_cvCalcArrBackProject in opencv2(histogram.o)

std::__1::vector<unsigned long, std::__1::allocator<unsigned long> >::__append(unsigned long) in opencv2(histogram.o)

std::__1::vector<double, std::__1::allocator<double> >::__append(unsigned long) in opencv2(histogram.o)

std::__1::vector<int, std::__1::allocator<int> >::__append(unsigned long) in opencv2(histogram.o)

std::__1::vector<unsigned char*, std::__1::allocator<unsigned char*> >::__append(unsigned long) in opencv2(histogram.o)

cv::SparseMat::resizeHashTab(unsigned long) in opencv2(matrix.o)

...

"std::__1::__basic_string_common<true>::__throw_length_error() const", referenced from:

std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> >::str() const in opencv2(ocl.o)

"std::__1::locale::use_facet(std::__1::locale::id&) const", referenced from:

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char const*) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, unsigned char) in opencv2(ocl.o)

"std::__1::ios_base::getloc() const", referenced from:

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char const*) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, unsigned char) in opencv2(ocl.o)

"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::resize(unsigned long, char)", referenced from:

std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> >:verflow(int) in opencv2(ocl.o)

std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> >::str(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) in opencv2(ocl.o)

"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >:ush_back(char)", referenced from:

std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> >:verflow(int) in opencv2(ocl.o)

"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::~basic_string()", referenced from:

cv:cl::kernelToStr(cv::_InputArray const&, int, char const*) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned short>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<short>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<int>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<float>(cv::Mat const&) in opencv2(ocl.o)

...

"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >:perator=(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&)", referenced from:

std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> >::str(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >::sentry::sentry(std::__1::basic_ostream<char, std::__1::char_traits<char> >&)", referenced from:

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char const*) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, unsigned char) in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >::sentry::~sentry()", referenced from:

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char const*) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, char) in opencv2(ocl.o)

std::__1::basic_ostream<char, std::__1::char_traits<char> >& std::__1:perator<<<std::__1::char_traits<char> >(std::__1::basic_ostream<char, std::__1::char_traits<char> >&, unsigned char) in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >::~basic_ostream()", referenced from:

construction vtable for std::__1::basic_ostream<char, std::__1::char_traits<char> >-in-std::__1::basic_ostringstream<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >::~basic_ostream()", referenced from:

construction vtable for std::__1::basic_ostream<char, std::__1::char_traits<char> >-in-std::__1::basic_ostringstream<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >::~basic_ostream()", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned short>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<short>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<int>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<float>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<double>(cv::Mat const&) in opencv2(ocl.o)

...

"std::__1::basic_ostream<char, std::__1::char_traits<char> >:perator<<(double)", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<double>(cv::Mat const&) in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >:perator<<(float)", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<float>(cv::Mat const&) in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >:perator<<(int)", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned short>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<short>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<int>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<float>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<double>(cv::Mat const&) in opencv2(ocl.o)

...

"std::__1::basic_ostream<char, std::__1::char_traits<char> >:perator<<(short)", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<short>(cv::Mat const&) in opencv2(ocl.o)

"std::__1::basic_ostream<char, std::__1::char_traits<char> >:perator<<(unsigned short)", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned short>(cv::Mat const&) in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::sync()", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::imbue(std::__1::locale const&)", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::uflow()", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::setbuf(char*, int)", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::xsgetn(char*, int)", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::xsputn(char const*, int)", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::showmanyc()", referenced from:

vtable for std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char> > in opencv2(ocl.o)

"std::__1::basic_streambuf<char, std::__1::char_traits<char> >::basic_streambuf()", referenced from:

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<unsigned char>(cv::Mat const&) in opencv2(ocl.o)

std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > cv:cl::kerToStr<char>(cv::Mat const&) in opencv2(ocl.o)

 

I was also verified in the same environment(Unity 4.5.5p1, XCode 6.1.1, OS X Yosemte 10.10.1), but the problem in the build did not occur.
opencv2.framework might not have been correctly added.

 

Released Version 1.0.8

Version changes
1.0.8
[Android]Support for x86 build target.(Unity 4.6 or higher)

 

Success to iOS 64bit build.

http://forum.unity3d.com/threads/4-6-ios-64-bit-beta.290551/
I have confirmed that the project using the "OpenCV ObjectDetector now version 1.0.8" in Unity4.6.1p3 is successful in iOS 64bit build.

 

Released Version 1.0.9

Version changes
1.0.9
[iOS]Support for arm64 build target.(Unity 4.6.1p3 or higher)

 

Released Version 1.1.0

Version changes
1.1.0
[Common]Update to OpenCV2.4.10

 

"OpenCV ObjectDetector 1.1.1(Support for Unity5)" is in pending review now in AssetStore.
I think that it is approved a few days later.

As for the current version1.1.0, only Android and Win and Mac support Unity5.
But, iOS does not support.
If you do not use iOS, there is not the problem with upgrading.

 

Released Version 1.1.1

Version changes
1.1.1
[Common]Support for Unity5.

 

Released Version 1.1.2

Version changes
1.1.2
[Common]Divide asset for Unity4 and Unity5.

 

Hey,
when using the Object Detector, is it possible to get the detected part of the texture into a new texture?
For example, look at this screen:

So that the detected face, which is marked blue, gets read into a new texture which just contains the part marked by your plugin?

Thanks in advance

 

Please use the Texture2D.GetPixels() and Texture2D.SetPixels().

Code (CSharp):

1. /// <summary>
2.         /// Simples the faces detect callback.
3.         /// </summary>
4.         /// <param name="result">Result.</param>
5.         void SimpleFacesDetectCallback (string result)
6.         {
7.                 Debug.Log ("SimpleFacesDetectCallback result" + result);
8.  
9.                 string json = result;
10.        
11.                 IDictionary detects = (IDictionary)Json.Deserialize (json);
12.        
13.                 foreach (DictionaryEntry detect in detects) {
14.                         Debug.Log ("detects key " + detect.Key);
15.            
16.                         string key = (string)detect.Key;
17.            
18.                         if (key.Equals ("error")) {
19.                                 Debug.Log ((string)detects [detect.Key]);
20.                         } else {
21.                
22.                                 IList<object> rects = (IList<object>)detects [detect.Key];
23.  
24.  
25.                                 //flip Rects by convenient method,
26.                                 IList<object> flipRects = OpenCVObjectDetector.FlipRects (rects, ((Texture2D)GetComponent<Renderer> ().material.mainTexture).width, ((Texture2D)GetComponent<Renderer> ().material.mainTexture).height, 0);
27.  
28.                                 Texture2D baseTexture = (Texture2D)GetComponent<Renderer> ().material.mainTexture;
29.                
30.                                 foreach (IDictionary rect in flipRects) {
31.                                         Texture2D newTexture = new Texture2D ((int)((long)rect ["width"]), (int)((long)rect ["height"]), TextureFormat.RGBA32, false);
32.                                         newTexture.SetPixels (baseTexture.GetPixels ((int)((long)rect ["x"]), (int)((long)rect ["y"]), (int)((long)rect ["width"]), (int)((long)rect ["height"])));
33.                                         newTexture.Apply ();
34.                    
35.                    
36.                                         gameObject.GetComponent<Renderer> ().material.mainTexture = newTexture;
37.                                 }
38.  
39.  
40. //                                #if UNITY_PRO_LICENSE || ((UNITY_ANDROID || UNITY_IPHONE) && !UNITY_EDITOR) || !(UNITY_4_5 || UNITY_4_6)
41. //                                OpenCVObjectDetector.DrawRects ((Texture2D)GetComponent<Renderer> ().material.mainTexture, Json.Serialize (flipRects), 0, 0, 255, 2);
42. //                                #endif
43. //
44. //
45. //                                ResultRectsToResultGameObjects (flipRects, new Color (0.0f, 0.0f, 1.0f, 0.3f), -40);
46.                                
47.                         }
48.                 }
49.         }

 

Thanks, got it working awesome. Now il just have the problem that i can only detect from textures integrated in the app. Since texture needs to me read/write enabled in the inspector. Is there a way around it? So i can define an texture 2d in runtime and use it with OpenCV Object Detector?

 

Since Texture2D that are generated from the code read and write is enabled, you can use in OpenCV Object Detector.

 

Hello,first of all GREAT PLUGIN!,but i can see the rects in the editor im on mac osx 10.10.3 with unity 5.0.1,i trace the "bug" and see that the plugin never enter in the callback,thank you in advance

 

Released Version 1.1.3

Version changes
1.1.3
[Common]Update to OpenCV2.4.11.

 

Released Version 1.1.4

Version changes
1.1.4
[Common]Add OpenCVObjectDetectorMenuItem.cs.( This script set plugin import settings automatically from MenuItem.)
[iOS]Move “OpenCVObjectDetector/ iOSforXcode/iOS_BuildPostprocessor.cs” to “OpenCVObjectDetector/Editor”folder.

 

Released Version 1.1.5

Version changes
1.1.5
[Common]Add SampleScene Setup Tutorial Video for Unity5.

 

ıt doesnt draw !!!!!!!!!!


ı put picture. where is the mistake ?

 

There is a possibility that the loading of xml file in the StreamingAssets folder has failed.
In Windows7, when I change the position of the project folder, there was a case to work well.

Please tell me the OS you are using.

 

Hi great plugin, however, using the ObjectDetectSample script provided and a Texture2d I am creating from webcam, the ResultCube seems to be instantiated in the wrong place, as in it is far from the image renderer not where the 2Dmarker shows up. Any suggestions as to why this might be happening?

 

Is the scale of Quad(GameObject) set to size of Texture2D?

 

I think the problem is that the Quad was a child of another object, thus its scale was being affected by its parent's scale.

 

Hi!
This plugin will be working with Android camera?
Thanks for answer.
Dmitry.

 

Hi.
I tried your face detect and it works perfectly, thanks.

I have a question, could you answer me please?
I get a cascade xml for detecting hand. Can I use it in the way you use your face cascade?

 

This asset supports only the detection of the object from Texture2D.

 

”OpenCVObjectDetector” is available with any cascade file.

 

Hi. Thanks for reply. Could you please answer 1 more question please...

I want to use OpenCV to compare 2 images. Could it be with this asset?

 

There is no function that this asset compares two images.
By using the "OpenCV for Unity", you can use the other functions of the OpenCV.
https://www.assetstore.unity3d.com/en/#!/content/21088
http://enoxsoftware.github.io/OpenC...mgproc.html#a958fbeec60087003c49d2297bfabcbc6

 

Thanks. I bought Object Detector, but "OpenCV for Unity" 's price is 95$ , i have to consider...

 

Released Version 1.1.6

Version changes
1.1.6
[iOS]Move “OpenCVObjectDetector/ iOSforXcode/opencv2.framework”to “OpenCVObjectDetector/Plugins/iOS/”folder.

 

Hi

I get a weird bug when using it!

When run it on 1 scene, it works perfectly. But every time i change scene, the call back not run! Until I restart Unity.

Could you help me please!

 

It even happen to your sample scene, when I switch to license scene then come back

 

Thank you very much for your attention.

As you said,This bug seems to happen at run time in the Editor.
To fix the problem, please fix the code as follows.

OnDisable() in OpenCVObjectDetector/Loom.cs

Code (CSharp):

1.     void OnDisable()
2.     {
3.         if (_current == this)
4.         {
5.          
6.             _current = null;
7.             initialized = false; // Insert this line.
8.         }
9.     }

I'm going to release a bug fix version as soon as possible.

 

Ah, thanks

 

Edited: Now it work when run plugin's tool Set Plugins Import Setting after import Admob. Note for anyone get same problem!

Sorry for any convenience, but I have a new problem!

When I get everything works fine, I integrate Admob plugin into the app. It still work fine on Editor, but cannot detect face on real Android phone.

Here is the Admob plugin i used: https://github.com/googleads/googleads-mobile-unity/tree/master/unity
It create something in Plugins/Android folder, with GoogleMobileAdsPlugin and google-play-services_lib as Android native plugins. Do them conflict with your plugins?

Thanks

 

Released Version 1.1.7

Version changes
1.1.7
[Common] Fixed the bug that occurs in the editor.

 

Hi there, I am interested in this plugin but I am just wondering how much weight it will add (roughly) to my app? Does it need the full 95 meg? Are there any strategies for reducing the size to the barest minimum necessary for just detecting faces? Or is this the smallest it can be using OpenCv?

 

Actually I saw there was a demo app that was 27 meg or so... thats better, but basically that is my target size for my entire app, is there anything that can be done to bring this down? Maybe some way of stripping stuff from OpenCV, doing a custom build of it or something, compressing the XML files etc etc

 

Unfortunately, there is no way to reduce the build size at present.

 

I feel you, its definately tricky. I am trying to do something at present, by converting things like trackingjs or liuliu's ccvjs to C#. These are teeny tiny little things that do what OpenCVObjectDetector does but with out the monolithic bulk, trackingjs is not that great but ccv seems pretty flawless (but it only detects faces, not eyes/nose/mouth)... If you'd be interested in collaborating on this PLEASE get in touch, I think a light weight face reco system could go barnstormers on the asset store, particularly with all the AR/VR stuff coming our way in 2016...

 

Released Version 1.1.8

Version changes
1.1.8
[iOS]Enabled Bitcode.