texture for YUV420 to RGB conversion in OpenGL ES - opengl-es

I have to convert and display YUV420P images to RGB colorspace using the AMD GPU on a Freescale iMX53 processor (OpenGL ES 2.0, EGL). Linux OS, no X11. To achieve this I should be able to create an appropriate image holding the YUV420P data: this could be either a YUV420P/YV12 image type or 3 simple 8-bit images, one for each component (Y, U, V).
glTexImage2D is excluded, because it's slow, the YUV420P frames are the results of a real time video decoding #25FPS and with glTexImage2D we can't keep the desired framerate.
There's an alternative: eglCreateImageKHR/glEGLImageTargetTexture2DOES. The only problem is that these can't handle any image format that would be suitable for YUV420/YV12 data.
EGLint attribs[] = {
EGL_WIDTH, 800,
EGL_HEIGHT, 480,
EGL_IMAGE_FORMAT_FSL, EGL_FORMAT_YUV_YV12_FSL,
EGL_NONE
};
EGLint const req_attribs[] = {
EGL_RED_SIZE, 5,
EGL_GREEN_SIZE, 6,
EGL_BLUE_SIZE, 5,
EGL_ALPHA_SIZE, 0,
EGL_SAMPLES, 0,
EGL_COLOR_BUFFER_TYPE, EGL_RGB_BUFFER,
EGL_SURFACE_TYPE, EGL_WINDOW_BIT,
EGL_NONE
};
...
display = eglGetDisplay(EGL_DEFAULT_DISPLAY);
eglInitialize(display, NULL, NULL);
eglBindAPI(EGL_OPENGL_ES_API);
eglChooseConfig(display, req_attribs, config, ARRAY_SIZE(config), &num_configs);
ctx = eglCreateContext(display, curr_config, NULL, NULL);
surface = eglCreateWindowSurface(display, curr_config, fb_handle, NULL);
...
EGLImageKHR yuv_img = eglCreateImageKHR(display, EGL_NO_CONTEXT, EGL_NEW_IMAGE_FSL, NULL, attribs);
eglQueryImageFSL(display, yuv_img, EGL_CLIENTBUFFER_TYPE_FSL, (EGLint *)&ptr);
glGenTextures(1, &texture);
glBindTexture(GL_TEXTURE_2D, texture);
glEGLImageTargetTexture2DOES(GL_TEXTURE_2D, yuv_img);
glEGLImageTargetTexture2DOES(...) fails. If I change the appropriate line in 'attribs' to this:
EGL_IMAGE_FORMAT_FSL, EGL_FORMAT_RGB_565_FSL,
then the image can be assigned to an OpenGL ES texture, but it's not appropriate to hold either 8-bit data (Y/U/V) or a YUV420/YV12 data. Searching the net (including Freescale community forum) I've haven't found any solution to this.
How can I create an image which:
is fast to create;
eventually can be assigned to an already existing buffer (physical address or virtual address is given);
can be used in the fragment/vertex shader program to perform YUV --> RGB conversion;
Constraint is to avoid unneccessary memcpy(...)s due to performance reasons.

I have implemented this on the i.MX53 for several YUV formats and it works really well. I have a published article about it, although it was generalized to cover more Android platforms:
http://software.intel.com/en-us/articles/using-opengl-es-to-accelerate-apps-with-legacy-2d-guis
I suspect your problem is that you are not binding to the correct texture target. It should be like this:
glEGLImageTargetTexture2DOES(GL_TEXTURE_EXTERNAL_OES, hEglImage[iTextureIndex]);
glBindTexture(GL_TEXTURE_EXTERNAL_OES, hTexture[iIndex]);
And the eglImageAttributes should be one of these:
EGLint eglImageAttributes[] = {EGL_WIDTH, iTextureWidth, EGL_HEIGHT, iTextureHeight, EGL_IMAGE_FORMAT_FSL, EGL_FORMAT_YUV_YV12_FSL, EGL_NONE};
EGLint eglImageAttributes[] = {EGL_WIDTH, iTextureWidth, EGL_HEIGHT, iTextureHeight, EGL_IMAGE_FORMAT_FSL, EGL_FORMAT_YUV_NV21_FSL, EGL_NONE};
EGLint eglImageAttributes[] = {EGL_WIDTH, iTextureWidth, EGL_HEIGHT, iTextureHeight, EGL_IMAGE_FORMAT_FSL, EGL_FORMAT_YUV_UYVY_FSL, EGL_NONE};
hEglImage[iTextureIndex] = eglCreateImageKHR(eglDisplay, EGL_NO_CONTEXT, EGL_NEW_IMAGE_FSL, NULL, eglImageAttributes);
struct EGLImageInfoFSL EglImageInfo;
eglQueryImageFSL(eglDisplay, hEglImage[iTextureIndex], EGL_CLIENTBUFFER_TYPE_FSL, (EGLint *)&EglImageInfo);
Although this feature of the Freescale i.MX53 platform makes YUV to RGB color space conversion for video extremely fast, it does have a couple of limitations:
It only supports those 3 YUV formats.
eglCreateImageKHR() must allocate the buffers. There is no way to make it use existing buffers. Freescale confirmed that the NULL pointer can not be anything else, which technically violates the Khronos specification.
Freescale has resolved these problems on the i.MX6 platform, although the architecture is really different. Hope this helps.

Related

AMD OpenGL and HDR display problem on Windows

I have been using OpenGL to display HDR content following explanation from nvidia:
https://on-demand.gputechconf.com/gtc/2017/presentation/s7394-tom-true-programming-for-high-dynamic-range.pdf
And it works great, but only on nVidia GPUs.
Using the same method:
Specify WGL_PIXEL_TYPE_ARB = WGL_TYPE_RGBA_FLOAT_ARB
with color depth 16 (WGL_RED_BITS_ARB = 16, WGL_GREEN_BITS_ARB = 16, WGL_BLUE_BITS_ARB = 16)
On AMD GPUs it displays SDR image.
That is to say it clamps the fragment shader values to 1.0, while on nvidia gpus it allows values to ~25.0 (or 10.000 nits as i understand it), and displays it correctly.
This is using the same TV (LG B9) and same OS (Wind 10).
Note that other apps, like Chrome displays HDR content correctly on AMD gpus, and directX tests apps also.
Tried bunch of different AMD GPUs, drivers settings texture formats, pixel types etc, with no luck.
Read thru whole https://gpuopen.com/ for clues, no luck.
Anyone have an idea or example how to create a proper OpenGL HDR Context/configuration?
I'll try an minimal example here, but its part of larger process and in Delphi, so it will be for orientation only
const
PixelaAttribList: array[0..20] of Integer =( //
WGL_DRAW_TO_WINDOW_ARB, 1, //
WGL_DOUBLE_BUFFER_ARB, 1, //
WGL_SUPPORT_OPENGL_ARB, 1, //
WGL_ACCELERATION_ARB, WGL_FULL_ACCELERATION_ARB, //
WGL_SWAP_METHOD_ARB, WGL_SWAP_EXCHANGE_ARB, //
WGL_PIXEL_TYPE_ARB, WGL_TYPE_RGBA_FLOAT_ARB, //
WGL_RED_BITS_ARB, 16, //
WGL_GREEN_BITS_ARB, 16, //
WGL_BLUE_BITS_ARB, 16, //
WGL_ALPHA_BITS_ARB, 0, //
0);
var
piFormats: PGLint;
Begin
wglChoosePixelFormatARB(DC, #PixelaAttribList, NIL, 100, piFormats, #nNumFormats);
if nNumFormats = 0 then
exit;
if not SetPixelFormat(DC, piFormats^, nil) then
exit;
hrc:= wglCreateContextAttribsARB(DC, 0, nil);
if Result <> 0 then
ActivateRenderingContext(DC, hrc);
After the code i tested format with
wglGetPixelFormatAttribivARB
and I get 16 bit per color, so exactly whats needed.
Fragment shader is simple:
gl_FragColor = vec4(25.0,25.0,25.0,1.0);
Regards

eglCreateImageKHR is working only with RGBA8888 format. How to send RGB888 data?

I implemented an opengl-es application running on mali-400 gpu. I grab the 1280x960 RGB buffer from camera and render on GPU , I use zero copy operation using EGL_EXT_image_dma_buf_import extension.
my problem is that the Besler camera gives frames in RGB888 format.
but eglCreateImageKHR is working only if I give RGBA8888 format . If I pass RGB888 to eglCreateImageKHR it gives black screen and also egl image returned is 0.
will I have to convert RGB888 to RGBA8888 before passing? what are the other options. below is my code.
EGLint egl_img_attr[] = { EGL_IMAGE_PRESERVED_KHR, EGL_TRUE,
EGL_DMA_BUF_PLANE0_FD_EXT, buffer->dbuf_fd,
EGL_DMA_BUF_PLANE0_OFFSET_EXT, 0,
EGL_DMA_BUF_PLANE0_PITCH_EXT, TEX_WIDTH * 3,
EGL_WIDTH, TEX_WIDTH,
EGL_HEIGHT, TEX_HEIGHT,
EGL_LINUX_DRM_FOURCC_EXT, DRM_FORMAT_RGBA8888 ,
EGL_NONE, EGL_NONE };
buffer->egl_img = eglCreateImageKHR(egl_dpy, EGL_NO_CONTEXT, EGL_LINUX_DMA_BUF_EXT, (EGLClientBuffer)0, egl_img_attr);

How to color manage AVAssetWriter output

I'm having trouble getting a rendered video's colors to match the source content's colors. I'm rendering images into a CGContext, converting the backing data into a CVPixelBuffer and appending that as a frame to an AVAssetWriterInputPixelBufferAdaptor. This causes slight color differences between the images that I'm drawing into the CGContext and the resulting video file.
It seems like there are 3 things that need to be addressed:
tell AVFoundation what colorspace the video is in.
make the AVAssetWriterInputPixelBufferAdaptor and the CVPixelBuffers I append to it match that color space.
use the same colorspace for the CGContext.
The documentation is terrible, so I'd appreciate any guidance on how to do these things or if there is something else I need to do to make the colors be preserved throughout this entire process.
Full code:
AVAssetWriter *_assetWriter;
AVAssetWriterInput *_assetInput;
AVAssetWriterInputPixelBufferAdaptor *_assetInputAdaptor;
NSDictionary *outputSettings = #{ AVVideoCodecKey :AVVideoCodecH264,
AVVideoWidthKey :#(outputWidth),
AVVideoHeightKey:#(outputHeight)};
_assetInput = [AVAssetWriterInput assetWriterInputWithMediaType:AVMediaTypeVideo
outputSettings:outputSettings];
NSDictionary *bufferAttributes = #{å(NSString*)kCVPixelBufferPixelFormatTypeKey:#(kCVPixelFormatType_32ARGB)};
_assetInputAdaptor = [AVAssetWriterInputPixelBufferAdaptor assetWriterInputPixelBufferAdaptorWithAssetWriterInput:_assetInput
sourcePixelBufferAttributes:bufferAttributes];
_assetWriter = [AVAssetWriter assetWriterWithURL:aURL fileType:AVFileTypeMPEG4 error:nil];
[_assetWriter addInput:_assetInput];
[_assetWriter startWriting];
[_assetWriter startSessionAtSourceTime:kCMTimeZero];
NSInteger bytesPerRow = outputWidth * 4;
long size = bytesPerRow * outputHeight;
CGColorSpaceRef srgbSpace = CGColorSpaceCreateWithName(kCGColorSpaceSRGB);
UInt8 *data = (UInt8 *)calloc(size, 1);
CGContextRef ctx = CGBitmapContextCreateWithData(data, outputWidth, outputHeight, 8, bytesPerRow, srgbSpace, kCGImageAlphaPremultipliedFirst, NULL, NULL);
// draw everything into ctx
CVPixelBufferRef pixelBuffer;
CVPixelBufferCreateWithBytes(kCFAllocatorSystemDefault,
outputWidth, outputHeight,
k32ARGBPixelFormat,
data,
bytesPerRow,
ReleaseCVPixelBufferForCVPixelBufferCreateWithBytes,
NULL,
NULL,
&pixelBuffer);
NSDictionary *pbAttachements = #{(id)kCVImageBufferCGColorSpaceKey : (__bridge id)srgbSpace};
CVBufferSetAttachments(pixelBuffer, (__bridge CFDictionaryRef)pbAttachements, kCVAttachmentMode_ShouldPropagate);
[_assetInputAdaptor appendPixelBuffer:pixelBuffer withPresentationTime:CMTimeMake(0, 60)];
CGColorSpaceRelease(srgbSpace);
[_assetInput markAsFinished];
[_assetWriter finishWritingWithCompletionHandler:^{}];
This is quite a confusing subject and the Apple docs really do not help all that much. I am going to describe the solution I have settled on based on using the BT.709 colorspace, I am sure someone will have an objection based on Colorimetric correctness and the weirdness of various video standards, but this is complex topic. First off, don't use kCVPixelFormatType_32ARGB as the pixel type. Always pass kCVPixelFormatType_32BGRA instead, since BGRA is the native pixel layout on both MacOSX and iPhone hardware and it BGRA is just faster. Next, when you create a CGBitmapContext to render into use the BT.709 colorspace (kCGColorSpaceITUR_709). Also, don't render into a malloc() buffer, render directly into the CoreVideo pixel buffer by creating a bitmap context over the same memory, CoreGraphics will handle the colorspace and gamma conversion from whatever your input image is to BT.709 and its associated gamma. Then you need to tell AVFoundation the colorspace of the pixel buffer, do that by making an ICC profile copy and setting the kCVImageBufferICCProfileKey on the CoreVideo pixel buffer. That takes care of your issues 1 and 2, you do not need to have input images in this same colorspace with this approach. Now, this is of course complex and actual working source code (yes actually working) is hard to come by. Here is a github link to a small project that does these exact steps, the code is BSD licensed, so feel free to use it. Note specifically the H264Encoder class which wraps all this horror up into a reusable module. You can find calling code in encode_h264.m, it is a little MacOSX command line util to encode PNG to M4V. Also attached 3 keys Apple docs related to this subject 1, 2, 3.
MetalBT709Decoder

Reading Pixels in WebGL 2 as Float values

I need to read the pixels of my framebuffer as float values.
My goal is to get a fast transfer of lots of particles between CPU and GPU and process them in realtime. For that I store the particle properties in a floating point texture.
Whenever a new particle is added, I want to get the current particle array back from the texture, add the new particle properties and then fit it back into the texture (this is the only way I could think of to dynamically add particles and process them GPU-wise).
I am using WebGL 2 since it supports reading back pixels to a PIXEL_PACK_BUFFER target. I test my code in Firefox Nightly. The code in question looks like this:
// Initialize the WebGLBuffer
this.m_particlePosBuffer = gl.createBuffer();
gl.bindBuffer(gl.PIXEL_PACK_BUFFER, this.m_particlePosBuffer);
gl.bindBuffer(gl.PIXEL_PACK_BUFFER, null);
...
// In the renderloop, bind the buffer and read back the pixels
gl.bindBuffer(gl.PIXEL_PACK_BUFFER, this.m_particlePosBuffer);
gl.readBuffer(gl.COLOR_ATTACHMENT0); // Framebuffer texture is bound to this attachment
gl.readPixels(0, 0, _texSize, _texSize, gl.RGBA, gl.FLOAT, 0);
I get this error in my console:
TypeError: Argument 7 of WebGLRenderingContext.readPixels could not be converted to any of: ArrayBufferView, SharedArrayBufferView.
But looking at the current WebGL 2 Specification, this function call should be possible. Using the type gl.UNSIGNED_BYTE also returns this error.
When I try to read the pixels in an ArrayBufferView (which I want to avoid since it seems to be way slower) it works with the format/type combination of gl.RGBA and gl.UNSIGNED_BYTE for a Uint8Array() but not with gl.RGBA and gl.FLOAT for a Float32Array() - this is as expected since it's documented in the WebGL Specification.
I am thankful for any suggestions on how to get my float pixel values from my framebuffer or on how to otherwise get this particle pipeline going.
Did you try using this extension?
var ext = gl.getExtension('EXT_color_buffer_float');
The gl you have is webgl1,not webgl2.Try:
var gl = document.getElementById("canvas").getContext('webgl2');
In WebGL2 the syntax for glReadPixel is
void gl.readPixels(x, y, width, height, format, type, ArrayBufferView pixels, GLuint dstOffset);
so
let data = new Uint8Array(gl.drawingBufferWidth * gl.drawingBufferHeight * 4);
gl.readPixels(0, 0, gl.drawingBufferWidth, gl.drawingBufferHeight, gl.RGBA, gl.UNSIGNED_BYTE, pixels, 0);
https://developer.mozilla.org/en-US/docs/Web/API/WebGLRenderingContext/readPixels

how to display data of yuv format without converting rgb in OpenGL ES?

I have being study about OpenGL ES for iOS.
I wonder that data of YUV format is can display without converting RGB.
In most, the yuv data have to convert RGB for display. But, converting process is very slow, Then, that is not display smoothly.
So, I would like to try to dispaly YUV data without convert to RGB.
Is it possible? If possible, what can I do?
Please, let me give a advice.
I think it is not possible in OpenGL ES to display YUV data without convert to RGB data.
You can do this very easily using OpenGL ES 2.0 shaders. I use this technique for my super-fast iOS camera app SnappyCam. The fragment shader would perform the matrix multiplication to take you from YCbCr ("YUV") to RGB. You can have each {Y, Cb, Cr} channel in a separate GL_LUMINANCE texture, or combine the {Cb, Cr} textures together using a GL_LUMINANCE_ALPHA texture if your chrominance data is already interleaved (Apple call this a bi-planar format).
See my related answer to the question YUV to RGBA on Apple A4, should I use shaders or NEON? here on StackOverflow.
You may also do this using the fixed rendering pipeline of ES 1.1, but I haven't tried it. I would look toward the the texture blending functions, e.g. given in this OpenGL Texture Combiners Wiki Page.
Is you are looking solution for IOS, iPhone application then there are solution for that.
This is way convert CMSampleBufferRef to UIImage when video pixel type is set as kCVPixelFormatType_420YpCbCr8BiPlanarFullRange
-(UIImage *) imageFromSamplePlanerPixelBuffer:(CMSampleBufferRef) sampleBuffer{
#autoreleasepool {
// Get a CMSampleBuffer's Core Video image buffer for the media data
CVImageBufferRef imageBuffer = CMSampleBufferGetImageBuffer(sampleBuffer);
// Lock the base address of the pixel buffer
CVPixelBufferLockBaseAddress(imageBuffer, 0);
// Get the number of bytes per row for the plane pixel buffer
void *baseAddress = CVPixelBufferGetBaseAddressOfPlane(imageBuffer, 0);
// Get the number of bytes per row for the plane pixel buffer
size_t bytesPerRow = CVPixelBufferGetBytesPerRowOfPlane(imageBuffer,0);
// Get the pixel buffer width and height
size_t width = CVPixelBufferGetWidth(imageBuffer);
size_t height = CVPixelBufferGetHeight(imageBuffer);
// Create a device-dependent gray color space
CGColorSpaceRef colorSpace = CGColorSpaceCreateDeviceGray();
// Create a bitmap graphics context with the sample buffer data
CGContextRef context = CGBitmapContextCreate(baseAddress, width, height, 8,
bytesPerRow, colorSpace, kCGImageAlphaNone);
// Create a Quartz image from the pixel data in the bitmap graphics context
CGImageRef quartzImage = CGBitmapContextCreateImage(context);
// Unlock the pixel buffer
CVPixelBufferUnlockBaseAddress(imageBuffer,0);
// Free up the context and color space
CGContextRelease(context);
CGColorSpaceRelease(colorSpace);
// Create an image object from the Quartz image
UIImage *image = [UIImage imageWithCGImage:quartzImage];
// Release the Quartz image
CGImageRelease(quartzImage);
return (image);
}
}
If you are looking for Mobile devices then i can provide others to.

Resources