c++ - DirectX Screen Capture and Output as Video

Question

I am doing desktop screen capture and wanted the output to be a video file. Currently I have codes getting from here to output png images. I modified the code a bit to change the output to JPEG file and then convert it to avi video output using openCV 3.0.0. Reason I need the JPEG file as output is because I am running on Windows 8.1 and OpenCV VideoWriter::fourcc('M', 'J', 'P', 'G') are the only options that work for me.

Output of PNG image works perfectly but not JPEG. There are vertical lines on the image and the time to generate the JPEG outputs is relatively far longer than PNG.

I have 2 options here:

1. Improve the output of JPEG file to work faster and have a clear image.

2. Get rid of the OpenCV 3.0.0 issue to accept PNG file input and able to output a video file.(preferably AVI/MP4 file format)

Either solution will do for me. PLease help. Thank you.

My code:

#include "stdafx.h"
#include <Wincodec.h>             // we use WIC for saving images
#include <d3d9.h>                 // DirectX 9 header
#include <opencvcv.h>
#include <opencvcxcore.hpp>
#include <opencv2highguihighgui.hpp>
#pragma comment(lib, "d3d9.lib")  // link to DirectX 9 library

using namespace cv;

#define WIDEN2(x) L ## x
#define WIDEN(x) WIDEN2(x)
#define __WFILE__ WIDEN(__FILE__)
#define HRCHECK(__expr) {hr=(__expr);if(FAILED(hr)){wprintf(L"FAILURE 0x%08X (%i)
line: %u file: '%s'
expr: '" WIDEN(#__expr) L"'
",hr, hr, __LINE__,__WFILE__);goto cleanup;}}
#define RELEASE(__p) {if(__p!=nullptr){__p->Release();__p=nullptr;}}

HRESULT Direct3D9TakeScreenshots(UINT adapter, UINT count);
HRESULT SavePixelsToFile32bppPBGRA(UINT width, UINT height, UINT stride, LPBYTE pixels, LPWSTR filePath, const GUID &format);


int _tmain(int argc, _TCHAR* argv[])
{
    HRESULT hr = Direct3D9TakeScreenshots(D3DADAPTER_DEFAULT, 10);
    return 0;
}


HRESULT Direct3D9TakeScreenshots(UINT adapter, UINT count)
{
    HRESULT hr = S_OK;
    IDirect3D9 *d3d = nullptr;
    IDirect3DDevice9 *device = nullptr;
    IDirect3DSurface9 *surface = nullptr;
    D3DPRESENT_PARAMETERS parameters = { 0 };
    D3DDISPLAYMODE mode;
    D3DLOCKED_RECT rc;
    UINT pitch;
    SYSTEMTIME st;
    LPBYTE *shots = nullptr;

    // init D3D and get screen size
    d3d = Direct3DCreate9(D3D_SDK_VERSION);
    HRCHECK(d3d->GetAdapterDisplayMode(adapter, &mode));

    parameters.Windowed = TRUE;
    parameters.BackBufferCount = 1;
    parameters.BackBufferHeight = mode.Height;
    parameters.BackBufferWidth = mode.Width;
    parameters.SwapEffect = D3DSWAPEFFECT_DISCARD;
    parameters.hDeviceWindow = NULL;

    // create device & capture surface
    HRCHECK(d3d->CreateDevice(adapter, D3DDEVTYPE_HAL, NULL, D3DCREATE_SOFTWARE_VERTEXPROCESSING, &parameters, &device));
    HRCHECK(device->CreateOffscreenPlainSurface(mode.Width, mode.Height, D3DFMT_A8R8G8B8, D3DPOOL_SYSTEMMEM, &surface, nullptr));

    // compute the required buffer size
    HRCHECK(surface->LockRect(&rc, NULL, 0));
    pitch = rc.Pitch;
    HRCHECK(surface->UnlockRect());

    // allocate screenshots buffers
    shots = new LPBYTE[count];
    for (UINT i = 0; i < count; i++)
    {
        shots[i] = new BYTE[pitch * mode.Height];
    }

    GetSystemTime(&st); // measure the time we spend doing <count> captures
    wprintf(L"START Capture--> %i:%i:%i.%i
", st.wHour, st.wMinute, st.wSecond, st.wMilliseconds);
    for (UINT i = 0; i < count; i++)
    {
        // get the data
        HRCHECK(device->GetFrontBufferData(0, surface));

        // copy it into our buffers
        HRCHECK(surface->LockRect(&rc, NULL, 0));
        CopyMemory(shots[i], rc.pBits, rc.Pitch * mode.Height);
        HRCHECK(surface->UnlockRect());
    }
    GetSystemTime(&st);
    wprintf(L"END Capture--> %i:%i:%i.%i
", st.wHour, st.wMinute, st.wSecond, st.wMilliseconds);

    // save all screenshots
    for (UINT i = 0; i < count; i++)
    {
        WCHAR file[100];
        wsprintf(file, L"cap%i.jpg", i);
        HRCHECK(SavePixelsToFile32bppPBGRA(mode.Width, mode.Height, pitch, shots[i], file, GUID_ContainerFormatJpeg));
    }

    cleanup:
    if (shots != nullptr)
    {
        for (UINT i = 0; i < count; i++)
        {
        delete shots[i];
        }
        delete[] shots;
    }

    RELEASE(surface);
    RELEASE(device);
    RELEASE(d3d);
    return hr;
}

HRESULT SavePixelsToFile32bppPBGRA(UINT width, UINT height, UINT stride, LPBYTE pixels, LPWSTR filePath, const GUID &format)
{
    if (!filePath || !pixels)
        return E_INVALIDARG;

    HRESULT hr = S_OK;
    IWICImagingFactory *factory = nullptr;
    IWICBitmapEncoder *encoder = nullptr;
    IWICBitmapFrameEncode *frame = nullptr;
    IWICStream *stream = nullptr;
    GUID pf = GUID_WICPixelFormat32bppPBGRA;
    BOOL coInit = CoInitialize(nullptr);

    HRCHECK(CoCreateInstance(CLSID_WICImagingFactory, nullptr, CLSCTX_INPROC_SERVER, IID_PPV_ARGS(&factory)));
    HRCHECK(factory->CreateStream(&stream));
    HRCHECK(stream->InitializeFromFilename(filePath, GENERIC_WRITE));
    HRCHECK(factory->CreateEncoder(format, nullptr, &encoder));
    HRCHECK(encoder->Initialize(stream, WICBitmapEncoderNoCache));
    HRCHECK(encoder->CreateNewFrame(&frame, nullptr)); // we don't use options here
    HRCHECK(frame->Initialize(nullptr)); // we dont' use any options here
    HRCHECK(frame->SetSize(width, height));
    HRCHECK(frame->SetPixelFormat(&pf));
    HRCHECK(frame->WritePixels(height, stride, stride * height, pixels));
    HRCHECK(frame->Commit());
    HRCHECK(encoder->Commit());

    cleanup:
    RELEASE(stream);
    RELEASE(frame);
    RELEASE(encoder);
    RELEASE(factory);
    if (coInit) CoUninitialize();

    //This part do encode JPEG file to video file 
    VideoCapture in_capture("cap%d.jpg");

    Mat img;

    VideoWriter out_capture("video.avi", CV_FOURCC('M','J','P','G'), 1, Size(1920,1080));

    while (true)
    {
        in_capture >> img;
        if(img.empty())
            break;

        out_capture.write(img);
    }

    return hr;
}

Answer 1

Your image isn't correct because the JPEG encoder encodes 3 formats JPEG Native Codec:

• GUID_WICPixelFormat8bppGray
• GUID_WICPixelFormat24bppBGR
• GUID_WICPixelFormat32bppCMYK

Did you try D3DXSaveSurfaceToFile, is it too slow ?

EDIT

For your second option, i've made a program to encode the screen capture with Media Foundation into h264 :

• This work on Windows Seven with 1280*1024 resolution, and with reasonable performance, but i don't known if it's feet your needs.
• This program is not real time capture, perhaps you will have to correct the timestamp passed to the sink writer. Consider using a Media Foundation clock.
• You can try to change encoding format : MF_TRANSCODE_CONTAINERTYPE
• Be careful that encoders are limited in resolution.
• Changing the MF_MT_AVG_BITRATE can improve quality over performance. Set the value according to your requirments.
• On my system, the capture image is reverse, so i need to copy the image in the rigth way (see #define REVERSE_IMAGE). Perhaps it is slower than MFCopyImage, but i'm not realy sure after testing both implementations.
• We don't need WIC, because Media Foundation encoders handle the format from the d3d9 capture.
• On Windows Seven, consider using IDirect3DDevice9Ex instead of idirect3ddevice9.
• I've mixed your d3d9 capture code with this tutorial : Using the Sink Writer to Encode Video. The design of the program can be improved too.

Here is the code :

#include <Windows.h>
#include <mfapi.h>
#include <mfidl.h>
#include <Mfreadwrite.h>
#include <mferror.h>
#include <d3d9.h>

#pragma comment(lib, "mfreadwrite")
#pragma comment(lib, "mfplat")
#pragma comment(lib, "mfuuid")
#pragma comment(lib, "d3d9.lib")

template <class T> void SafeRelease(T **ppT){

    if(*ppT){
        (*ppT)->Release();
        *ppT = NULL;
    }
}

#define REVERSE_IMAGE

// Format constants
const UINT32 VIDEO_FPS = 30;
const UINT64 VIDEO_FRAME_DURATION = 10 * 1000 * 1000 / VIDEO_FPS;
const UINT32 VIDEO_BIT_RATE = 2000000;
const GUID   VIDEO_ENCODING_FORMAT = MFVideoFormat_H264;
const GUID   VIDEO_INPUT_FORMAT = MFVideoFormat_RGB32;
const UINT32 VIDEO_FRAME_COUNT = 5 * VIDEO_FPS;

HRESULT InitializeDirect3D9(IDirect3DDevice9** ppDevice, IDirect3DSurface9** ppSurface, UINT32& uiWidth, UINT32& uiHeight){

    IDirect3D9* d3d = NULL;

    d3d = Direct3DCreate9(D3D_SDK_VERSION);

    if(d3d == NULL)
        return E_POINTER;

    D3DDISPLAYMODE mode;
    HRESULT hr = d3d->GetAdapterDisplayMode(D3DADAPTER_DEFAULT, &mode);

    if(FAILED(hr)){
        SafeRelease(&d3d);
        return hr;
    }

    D3DPRESENT_PARAMETERS parameters = {0};

    parameters.Windowed = TRUE;
    parameters.BackBufferCount = 1;
    uiHeight = parameters.BackBufferHeight = mode.Height;
    uiWidth = parameters.BackBufferWidth = mode.Width;
    parameters.SwapEffect = D3DSWAPEFFECT_DISCARD;
    parameters.hDeviceWindow = NULL;

    hr = d3d->CreateDevice(D3DADAPTER_DEFAULT, D3DDEVTYPE_HAL, NULL, D3DCREATE_SOFTWARE_VERTEXPROCESSING, &parameters, ppDevice);

    if(FAILED(hr)){
        SafeRelease(&d3d);
        return hr;
    }

    hr = (*ppDevice)->CreateOffscreenPlainSurface(mode.Width, mode.Height, D3DFMT_A8R8G8B8, D3DPOOL_SYSTEMMEM, ppSurface, nullptr);

    SafeRelease(&d3d);

    return hr;
}

HRESULT InitializeSinkWriter(IMFSinkWriter **ppWriter, DWORD *pStreamIndex, const UINT32 uiWidth, const UINT32 uiHeight){

    *ppWriter = NULL;
    *pStreamIndex = NULL;

    IMFSinkWriter   *pSinkWriter = NULL;
    IMFMediaType    *pMediaTypeOut = NULL;
    IMFMediaType    *pMediaTypeIn = NULL;
    DWORD           streamIndex;

    HRESULT hr = MFCreateSinkWriterFromURL(L"output.mp4", NULL, NULL, &pSinkWriter);

    // Set the output media type.
    if(SUCCEEDED(hr)){
        hr = MFCreateMediaType(&pMediaTypeOut);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeOut->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Video);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeOut->SetGUID(MF_MT_SUBTYPE, VIDEO_ENCODING_FORMAT);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeOut->SetUINT32(MF_MT_AVG_BITRATE, VIDEO_BIT_RATE);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeOut->SetUINT32(MF_MT_INTERLACE_MODE, MFVideoInterlace_Progressive);
    }
    if(SUCCEEDED(hr)){
        hr = MFSetAttributeSize(pMediaTypeOut, MF_MT_FRAME_SIZE, uiWidth, uiHeight);
    }
    if(SUCCEEDED(hr)){
        hr = MFSetAttributeRatio(pMediaTypeOut, MF_MT_FRAME_RATE, VIDEO_FPS, 1);
    }
    if(SUCCEEDED(hr)){
        hr = MFSetAttributeRatio(pMediaTypeOut, MF_MT_PIXEL_ASPECT_RATIO, 1, 1);
    }
    if(SUCCEEDED(hr)){
        hr = pSinkWriter->AddStream(pMediaTypeOut, &streamIndex);
    }

    // Set the input media type.
    if(SUCCEEDED(hr)){
        hr = MFCreateMediaType(&pMediaTypeIn);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeIn->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Video);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeIn->SetGUID(MF_MT_SUBTYPE, VIDEO_INPUT_FORMAT);
    }
    if(SUCCEEDED(hr)){
        hr = pMediaTypeIn->SetUINT32(MF_MT_INTERLACE_MODE, MFVideoInterlace_Progressive);
    }
    if(SUCCEEDED(hr)){
        hr = MFSetAttributeSize(pMediaTypeIn, MF_MT_FRAME_SIZE, uiWidth, uiHeight);
    }
    if(SUCCEEDED(hr)){
        hr = MFSetAttributeRatio(pMediaTypeIn, MF_MT_FRAME_RATE, VIDEO_FPS, 1);
    }
    if(SUCCEEDED(hr)){
        hr = MFSetAttributeRatio(pMediaTypeIn, MF_MT_PIXEL_ASPECT_RATIO, 1, 1);
    }
    if(SUCCEEDED(hr)){
        hr = pSinkWriter->SetInputMediaType(streamIndex, pMediaTypeIn, NULL);
    }

    // Tell the sink writer to start accepting data.
    if(SUCCEEDED(hr)){
        hr = pSinkWriter->BeginWriting();
    }

    // Return the pointer to the caller.
    if(SUCCEEDED(hr)){

        *ppWriter = pSinkWriter;
        (*ppWriter)->AddRef();
        *pStreamIndex = streamIndex;
    }

    SafeRelease(&pSinkWriter);
    SafeRelease(&pMediaTypeOut);
    SafeRelease(&pMediaTypeIn);
    return hr;
}

HRESULT WriteFrame(IDirect3DDevice9* pDevice, IDirect3DSurface9* pSurface, IMFSinkWriter* pWriter, DWORD streamIndex, const LONGLONG& rtStart, const UINT32 uiWidth, const UINT32 uiHeight){

    HRESULT hr = pDevice->GetFrontBufferData(0, pSurface);

    if(FAILED(hr)){
        return hr;
    }

    D3DLOCKED_RECT rc;
    hr = pSurface->LockRect(&rc, NULL, 0);

    if(FAILED(hr)){
        return hr;
    }

    IMFSample *pSample = NULL;
    IMFMediaBuffer *pBuffer = NULL;

    const LONG cbWidth = 4 * uiWidth;
    const DWORD cbBuffer = cbWidth * uiHeight;

    BYTE *pData = NULL;

    // Create a new memory buffer.
    hr = MFCreateMemoryBuffer(cbBuffer, &pBuffer);

    // Lock the buffer and copy the video frame to the buffer.
    if(SUCCEEDED(hr)){
        hr = pBuffer->Lock(&pData, NULL, NULL);
    }

    if(SUCCEEDED(hr)){

#ifdef REVERSE_IMAGE
        for(int i = 0, j = uiHeight - 1; i < uiHeight; i++, j--)
            for(int k = 0; k < cbWidth; k++)
                    pData[(i * cbWidth) + k] = ((BYTE*)rc.pBits)[(j * cbWidth) + k];
#else
        hr = MFCopyImage(pData, cbWidth, (BYTE*)rc.pBits, rc.Pitch, cbWidth, uiHeight);
#endif
    }

    if(pBuffer){
        pBuffer->Unlock();
    }

    // Set the data length of the buffer.
    if(SUCCEEDED(hr)){
        hr = pBuffer->SetCurrentLength(cbBuffer);
    }

    // Create a media sample and add the buffer to the sample.
    if(SUCCEEDED(hr)){
        hr = MFCreateSample(&pSample);
    }

    if(SUCCEEDED(hr)){
        hr = pSample->AddBuffer(pBuffer);
    }

    // Set the time stamp and the duration.
    if(SUCCEEDED(hr)){
        hr = pSample->SetSampleTime(rtStart);
    }

    if(SUCCEEDED(hr)){
        hr = pSample->SetSampleDuration(VIDEO_FRAME_DURATION);
    }

    // Send the sample to the Sink Writer.
    if(SUCCEEDED(hr)){
        hr = pWriter->WriteSample(streamIndex, pSample);
    }

    hr = pSurface->UnlockRect();

    SafeRelease(&pSample);
    SafeRelease(&pBuffer);
    return hr;
}

void main(){

    HRESULT hr = CoInitializeEx(NULL, COINIT_APARTMENTTHREADED);

    if(SUCCEEDED(hr)){

        hr = MFStartup(MF_VERSION);

        if(SUCCEEDED(hr)){

                UINT32 uiWidth = 0;
                UINT32 uiHeight = 0;

                IDirect3DDevice9* pDevice = NULL;
                IDirect3DSurface9* pSurface = NULL;

                hr = InitializeDirect3D9(&pDevice, &pSurface, uiWidth, uiHeight);

                if(SUCCEEDED(hr)){

                        IMFSinkWriter *pSinkWriter = NULL;
                        DWORD stream;

                        hr = InitializeSinkWriter(&pSinkWriter, &stream, uiWidth, uiHeight);

                        if(SUCCEEDED(hr)){

                            LONGLONG rtStart = 0;

                            for(DWORD i = 0; i < VIDEO_FRAME_COUNT; ++i){

                                hr = WriteFrame(pDevice, pSurface, pSinkWriter, stream, rtStart, uiWidth, uiHeight);

                                            if(FAILED(hr)){
                                                    break;
                                            }

                                            rtStart += VIDEO_FRAME_DURATION;
                                    }
                            }

                            if(SUCCEEDED(hr)){
                                    hr = pSinkWriter->Finalize();
                            }

                            SafeRelease(&pSinkWriter);
                    }

                    SafeRelease(&pDevice);
                    SafeRelease(&pSurface);
                    MFShutdown();
            }

            CoUninitialize();
    }
}