So I'm getting frames in CV420YpCbCr8BiPlanarFullRange pixel format. That is a bi-planar 4:2:0 thingy.
I need to convert it to some of the formats consumed by a third party library, for example this one.
At first sight it seems that it's the same format, but unfortunately the picture is mangled. Any ideas where is the problem? And what are some libraries that allow for conversion between the formats?
video_frame.p_data = (uint8_t *)CVPixelBufferGetBaseAddressOfPlane(pixelBuffer, 1);
doesn't work. Neither does getting the base address of the structure.
Related
There's no support TGA format for OpenCV currently.
And I know there's a single header file library named stb_image that allow you to read/write TGA image.
But the use case with OpenCV on the Internet are so few. (more often to see people use it with OpenGL)
The second method I found.
There's a short code included (the answer) in this topic:
Loading a tga/bmp file in C++/OpenGL
Someone use this code to read TGA file into cv::Mat just like the code below.
Tga tgaImg = Tga("/tmp/test.tga");
Mat img(tgaImg.GetHeight(), tgaImg.GetWidth(), CV_8UC4);
memcpy(img.data, tgaImg.GetPixels().data(), tgaImg.GetHeight() * tgaImg.GetWidth() * 4);
But this is only for reading part. I wonder if stb_image can do the same thing like the code above. I mean the image data structure might be different. (not look into them yet)
I would like to ask people who also experience this before. Since DDS/TGA image format are also popular using in game texture, there must be people have already found the way. I mean read/write TGA format in OpenCV code.
Thanks.
For saving opencv image in tga use stbi_write_tga. This function takes pointer to image data as argument, which is img.data in case of cv::Mat type.
I am not familiar with directx, but I ran into a problem in a small project, part of which involves capturing directx data. I hope, below I make some sense.
General question:
I would like to know what factors determine the DXGI_FORMAT of a texture in the backbuffer (hardware?, OS?, application?, directx version?). And more importantly, when capturing a texture from the backbuffer, is it possible to receive a texture in the desired format by supplying the format as a parameter, having the format automatically converted if necessary.
Specifics about my problem :
I capture screens from games using Open Broadcaster Software(OBS) and process them using a specific library(OpenCV) prior to streaming. I noticed that, following updates to both Windows and OBS, I get 'DXGI_FORMAT_R10G10B10A2_UNORM' as the DXGI_FORMAT. This is a problem for me, because as far as I know OpenCV does not provide a convenient way for building an OpenCV object when colors are 10bits. Below are a few relevant lines from the modified OBS source file.
d3d11_copy_texture(data.texture, backbuffer);
...
hlog(toStr(data.format)); // prints 24 = DXGI_FORMAT_R10G10B10A2_UNORM
...
ID3D11Texture2D* tex;
bool success = create_d3d11_stage_surface(&tex);
if (success) {
...
HRESULT hr = data.context->Map(tex, subresource, D3D11_MAP_READ, 0, &mappedTex);
...
Mat frame(data.cy, data.cx, CV_8UC4, mappedTex.pData, (int)mappedTex.RowPitch); //This creates an OpenCV Mat object.
//No support for 10-bit coors. Expects 8-bit colors (CV_8UC4 argument).
//When the resulting Mat is viewed, colours are jumbled (Probably because 10-bits did not fit into 8-bits).
Before the updates (when I was working on this a year ago), I was probably receiving DXGI_FORMAT = DXGI_FORMAT_B8G8R8A8_UNORM, because the code above used to work.
Now I wonder what changed, and whether I can modify the source code of OBS to receive data with the desired DXGI_FORMAT.
'create_d3d11_stage_surface' method called above sets the DXGI_FORMAT, but I am not sure if it means 'give me data with this DXGI_FORMAT' or 'I know you work with this format, give me what you have'.
static bool create_d3d11_stage_surface(ID3D11Texture2D **tex)
{
HRESULT hr;
D3D11_TEXTURE2D_DESC desc = {};
desc.Width = data.cx;
desc.Height = data.cy;
desc.Format = data.format;
...
I hoped that, overriding the desc.Format with DXGI_FORMAT_B8G8R8A8_UNORM would result in that format being passed as argument in the ID3D11DeviceContext::Map call above, and I would get data with specified format. But that did not work.
The choice of render target is up to the application, but they need to pick one based on the Direct3D hardware feature level. Formats for render targets in swapchains are usually display scanout formats:
DXGI_FORMAT_R8G8B8A8_UNORM
DXGI_FORMAT_R8G8B8A8_UNORM_SRGB
DXGI_FORMAT_B8G8R8A8_UNORM
DXGI_FORMAT_B8G8R8A8_UNORM
DXGI_FORMAT_R10G10B10A2_UNORM
DXGI_FORMAT_R16G16B16A16_FLOAT
DXGI_FORMAT_R10G10B10_XR_BIAS_A2_UNORM (rare)
See the DXGI documentation for the full list of supported formats and usages by feature level.
Direct3D 11 does not do format conversions when you copy resources such as copying to staging render textures, so if you want to do a format conversion you'll need to handle that yourself. Note that CPU-side conversion code for all the DXGI formats can be found in DirectXTex.
It is the application that decides that format. The simplest one would be R8G8B8A8, which simply represents RGB and alpha values. But, if developer decides that he will be using HDR, the backbuffer would probably be R11B11G10, because you can store way more precise data there, without alpha channel information. If the game is for example black and white, there's no need to keep all RGB channels in the back buffer, you could use simpler format. I hope this helps.
I have a 16bit grayscale image. I have tried both .png and .tif. .tif works somewhat. I have the following code:
CGDataProviderRef l_Img_Data_Provider = CGDataProviderCreateWithFilename( [m_Name cStringUsingEncoding:NSASCIIStringEncoding] );
CGImageRef l_CGImageRef = CGImageCreate( m_Width, m_Height, 16, 16, m_Width * 2,
CGColorSpaceCreateDeviceGray(), kCGBitmapByteOrder16Big, l_Img_Data_Provider, NULL, false, kCGRenderingIntentDefault );
test_Image = [[UIImage alloc] initWithCGImage:l_CGImageRef];
[_test_Image_View setImage:test_Image];
This results in the following image:
faulty gradient
As you can see, there seems to be an issue at the beginning of the image ( could it be trying to use the byte data from the header? ), and the image is offset by about a fifth ( a little harder to see, look at the left and the right, there is a faint line about a fifth away from the right.
My goal is to convert this to a metal texture and use it from there. Also having issues there. Seem like a byte order issue but maybe we can come back to that.
dave
CGDataProvider doesn't know about the format of the data that it stores. It is just meant for handling generic data:
"The CGDataProvider header file declares a data type that supplies
Quartz functions with data. Data provider objects abstract the
data-access task and eliminate the need for applications to manage
data through a raw memory buffer."
CGDataProvider
Because CGDataProvider is generic you must provide the format of the image data using the CGImageCreate parameters. PNGs and JPGs have their own CGImageCreateWith.. functions for handling encoded data.
The CGImage parameters in your example correctly describe a 16 bit grayscale raw byte format but nothing about TIF encoding so I would guess you are correct in guessing that the corrupted pixels you are see are from the file headers.
There may be other ways to load a 16 bit grayscale image on iOS, but to use that method (or the very similar Metal method) you would need to parse the image bytes from the TIF file and pass that into the function, or create another way to store and parse the image data.
My current problem is that I would like to know the type of the cv::Mat-frames grabbed by cv::VideoCapture from a video file. The documentation doesn't specify that (as is often the case, so even if I have overlooked it in this particular case, it would still be helpful to get an answer for dealing with the problem in general).
Of course, I could open the appropriate OpenCV header file and go through the CV_64FC2, ... macros to find a macro which matches the Mat's type(). But I'm kind of sick of that. There must be an easier way.
Isn't there any function that lets me translate a Mat's type() to a human-readable format ? Like this:
cv::Mat myMatWithUnknownType;
// Some code modifying myMatWithUnknownType.
// ...
std::string readableType = myMatWithUnknownType.typeString();
std::cout << readableType; // Prints "CV_64FC3".
How do you deal with that?
First, the format that come from cameras is only one: CV_8UC3. This is hardcoded in OpenCV, and any video format is converted to this before being sent to user. So
capture >> frame;
Will always return a RGB image, of 8 bits per channel.
Now, for other types you can write your function, keeping in mid that there are not so many types supported in OpenCV: A Mat can be of type char, uchar, short, ushort, int, uint, float, double, of 1 to 512 channels (according to the latest docs.) So writing your own type_to_string() is not difficult.
I'm having a proprietary image format SNG( a proprietary format) which is having a countinous array of Image data along with Image meta information in seperate HDR file.
Now I need to convert this SNG format to a Standard TIFF 6.0 Format. So I studied the TIFF format i.e. about its Header, Image File Directories( IFD's) and Stripped Image Data.
Now I have few concerns about this conversion. Please assist me.
SNG Continous Data vs TIFF Stripped Data: Should I convert SNG Data to TIFF as a continous data in one Strip( data load/edit time problem?) OR make logical StripOffsets of the SNG Image data.
SNG Data Header uses only necessary Meta Information, thus while converting the SNG to TIFF, some information can’t be retrieved such as NewSubFileType, Software Tag etc.
So this raises a concern that after conversion whether any missing directory information such as NewSubFileType, Software Tag etc is necessary and sufficient condition for TIFF File.
Encoding of each pixel component of RGB Sample in SNG data:
Here each SNG Image Data Strip per Pixel component is encoded as:
Out^[i] := round( LineBuffer^[i * 3] * **0.072169** + LineBuffer^[i * 3 + 1] * **0.715160** + LineBuffer^[i * 3+ 2]* **0.212671**);
Only way I deduce from it is that each Pixel is represented with 3 RGB component and some coefficient is multiplied with each component to make the SNG Viewer work RGB color information of SNG Image Data. (Developer who earlier work on this left, now i am following the trace :))
Thus while converting this to TIFF, the decoding the same needs to be done. This raises a concern that the how RBG information in TIFF is produced, or better do we need this information?.
Please assist...
Are you able to load this into a standard windows bitmap handle? If so, there are probably a bunch of free and commercial libraries for saving it as TIFF.
The standard for TIFF is libtiff -- it's a C library. Here's a version for Delphi made by an expert in the TIFF format:
http://www.awaresystems.be/imaging/tiff/delphi.html
There seems to be a lot of choices.
I think the approach of
Loading your format into an in-memory standard bitmap (which you need to do to show it, right?)
Using a pre-existing TIFF encoding library to save as TIFF
Will be a lot easier than trying to do a direct format-to-format conversion. The only reasons I wouldn't do it this way are:
The bitmap is too big to keep in memory
The original format is lossy and I will lose more quality in the re-encoding -- but you'd have to be saving in a standard lossy format (JPEG) to save quality.
Disclaimer: I work for Atalasoft.
We make .NET imaging codecs (including TIFF) -- that are a lot easier to use than LibTiff -- you can call them in Delphi through COM. We can convert standard windows bitmaps to TIFF or PDF (or other formats) with a couple of lines of code.
One approach, if you have a Windows application which handles and can print this format, would be to let it do the work for you, and call it to print the file to one of the many available 'printer drivers' which support direct output to TIFF.