kidapu

Any updates ?? I think OpenGLES is very slow to create AR App.

I want to use ARCore with Vulkan, too. Any updates?

I updated `Android.mk` and rebuilt `libRenderingPlugin.so`. ``` Android.mk APP_OPTIM := debug APP_ABI := armeabi-v7a arm64-v8a APP_PLATFORM := android-21 # APP_STL := gnustl_static APP_STL := c++_static APP_CPPFLAGS += -std=c++11 NDK_TOOLCHAIN_VERSION :=...

When I run `glGetError()` before assert(), I got this view on Android. Maybe only vertex shader is disable.  And I printf( glGetError() ), it says `GL_INVALID_ENUM` ( int 1280)....

I forked your repository, and add some code for investigating this bug. https://github.com/kidapu/NativeRenderingPlugin/commits/master

@genekogan I cloned latest darknet repository, and build it, and replace ofxDarknet in forked my repository. https://github.com/kidapu/ofxDarknet/commit/4434d9a3c4199f6e5ae9de42c52ac8eb48e30b52 I encountered many errors, I fixed it as much as possible. However, I...

@prabindh I use `nvidia/cuda:8.0-devel-ubuntu16.04` from this Dockerfile. https://gitlab.com/nvidia/cuda/blob/ubuntu16.04/8.0/devel/cudnn5/Dockerfile#L1

I have re-trained my face data by CuDNN =1 once. Following graph shows my train log, (x,y) = (epoch, loss rate). I have tried 29000 epochs. My CuDNN version is...

@bobeo No. I have not solved. Completely same happens to me!!!

In summary, the following happens in my case. (1) CuDNN == 0 && ( darknet-cpp || darknet-cpp-shared) coco & my dataset works fine. (2) CuDNN == 1 && ( darknet-cpp...