In my code, i am making 2 rectangles,
Rectangle1: Rendering On a Texture.
Rectangle2: Rendering On Back Buffer.
I am trying to do programmable blending,so need to access the destination pixel in pixel shader.
In my code,
I am creating a texture like below:
d3d11Device->CreateTexture2D(&textureDesc, NULL, &renderTargetTextureMap);
After this i am creating Render Target view of the texture.
d3d11Device->CreateRenderTargetView(renderTargetTextureMap, &renderTargetViewDesc, &renderTargetViewMap);
After this i am declaring vertex and pixel shader.
Then in my draw call,
i am performing following thing:
float bgColor[4] = {0.0f, 0.0f,0.0f, 1.0f };
d3d11DevCon->ClearRenderTargetView(renderTargetViewMap, bgColor);
float bgColor2[4] = { 0.0f, 1.0f, 0.0f, 1.0f };
////////////////////////////////////////////////Buffer 1///////////////////////////////////////////////////////////////
//Set the vertex buffer
UINT stride = sizeof(Vertex);
UINT offset = 0;
///////////////////////////////////////////////////////////Buffer 2//////////////////////////////////////////////////////////////////
d3d11DevCon->IASetIndexBuffer(d2dIndexBuffer, DXGI_FORMAT_R32_UINT, 0);
d3d11DevCon->IASetVertexBuffers(0, 1, &triangleVertBuffer, &stride, &offset);
////Draw the triangle
d3d11DevCon->DrawIndexed(6, 0, 0);
I assume that since i have set my render target view as renderTargetViewMap, so my draw call will render to texture only.
Now i am rendering to my backbuffer:
////////////////////////////////////////////
d3d11DevCon->OMSetRenderTargets(1, &renderTargetView, NULL);
d3d11DevCon->PSSetShaderResources(0, 1, &shaderResourceViewMap);
//d3d11DevCon->ClearRenderTargetView(renderTargetView, bgColor2);
d3d11DevCon->IASetIndexBuffer(d2dIndexBuffer2, DXGI_FORMAT_R32_UINT, 0);
d3d11DevCon->IASetVertexBuffers(0, 1, &triangleVertBuffer2, &stride, &offset);
////Draw the triangle
d3d11DevCon->DrawIndexed(6, 0, 0);
//Present the backbuffer to the screen
SwapChain->Present(0, 0);
So, in this way my rendering is happening.
Issue Face:
In my pixel shader,
VS_OUTPUT VS(float4 inPos : POSITION, float4 inColor : COLOR)
{
VS_OUTPUT output;
output.Pos = inPos;
output.Color = inColor;
return output;
}
float4 PS(VS_OUTPUT input) : SV_TARGET
{
float2 temp;
temp = input.Pos;
float4 diffuse = ObjTexture.Sample(ObjSamplerState,0.5+0.5*temp);
return input.Color + diffuse;
}
Here the diffuse is comming out to be equal to my bgcolor which i have set when rendering to texture
float bgColor[4] = {0.0f, 0.0f,0.0f, 1.0f };
d3d11DevCon->ClearRenderTargetView(renderTargetViewMap, bgColor);
I have also drawn a rectangle on it, but those pixels i am not able to access.
How can i access the pixel of rectangle that i have drawn on rendering to texture.
This is Issue Image
Desired Result
Shader File: Effect.fx
struct VS_OUTPUT
{
float4 Pos : SV_POSITION;
float4 Color : COLOR;
};
Texture2D ObjTexture;
SamplerState ObjSamplerState;
VS_OUTPUT VS(float4 inPos : POSITION, float4 inColor : COLOR)
{
VS_OUTPUT output;
output.Pos = inPos;
output.Color = inColor;
return output;
}
float4 PS(VS_OUTPUT input) : SV_TARGET
{
float2 temp;
temp = input.Pos;
float4 diffuse = ObjTexture.Sample(ObjSamplerState,0.5+0.5*temp);
return input.Color + diffuse;
}
Edit - 1:
With my latest change in code, i am able to blend my rectangle 2 with rectangle 1, but one issue i am facing is that when i blend then major part of my rectangle 2 is changes to yellow (red + green) on edges only i am able to see the actual green color.
Modified code:
main.cpp
//Include and link appropriate libraries and headers//
#pragma comment(lib, "d3d11.lib")
#pragma comment(lib, "d3dx11.lib")
#pragma comment(lib, "d3dx10.lib")
#include <windows.h>
#include <d3d11.h>
#include <d3dx11.h>
#include <d3dx10.h>
#include <xnamath.h>
//Global Declarations - Interfaces//
IDXGISwapChain* SwapChain;
ID3D11Device* d3d11Device;
ID3D11DeviceContext* d3d11DevCon;
ID3D11RenderTargetView* renderTargetView;
ID3D11Buffer* triangleVertBuffer;
ID3D11Buffer* triangleVertBuffer2;
ID3D11VertexShader* VS;
ID3D11PixelShader* PS;
ID3D10Blob* VS_Buffer;
ID3D10Blob* PS_Buffer;
ID3D11InputLayout* vertLayout;
XMMATRIX mapView;
XMMATRIX mapProjection;
XMVECTOR DefaultForward = XMVectorSet(0.0f, 0.0f, 1.0f, 0.0f);
//Global Declarations - Others//
LPCTSTR WndClassName = L"firstwindow";
HWND hwnd = NULL;
HRESULT hr;
const int Width = 800;
const int Height = 600;
bool InitializeDirect3d11App(HINSTANCE hInstance)
{
//Describe our Buffer
DXGI_MODE_DESC bufferDesc;
ZeroMemory(&bufferDesc, sizeof(DXGI_MODE_DESC));
bufferDesc.Width = Width;
bufferDesc.Height = Height;
bufferDesc.RefreshRate.Numerator = 60;
bufferDesc.RefreshRate.Denominator = 1;
bufferDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
bufferDesc.ScanlineOrdering = DXGI_MODE_SCANLINE_ORDER_UNSPECIFIED;
bufferDesc.Scaling = DXGI_MODE_SCALING_UNSPECIFIED;
//Describe our SwapChain
DXGI_SWAP_CHAIN_DESC swapChainDesc;
ZeroMemory(&swapChainDesc, sizeof(DXGI_SWAP_CHAIN_DESC));
swapChainDesc.BufferDesc = bufferDesc;
swapChainDesc.SampleDesc.Count = 1;
swapChainDesc.SampleDesc.Quality = 0;
swapChainDesc.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT;
swapChainDesc.BufferCount = 1;
swapChainDesc.OutputWindow = hwnd;
swapChainDesc.Windowed = TRUE;
swapChainDesc.SwapEffect = DXGI_SWAP_EFFECT_DISCARD;
//Create our SwapChain
hr = D3D11CreateDeviceAndSwapChain(NULL, D3D_DRIVER_TYPE_HARDWARE, NULL, NULL, NULL, NULL,
D3D11_SDK_VERSION, &swapChainDesc, &SwapChain, &d3d11Device, NULL, &d3d11DevCon);
//Create our BackBuffer
ID3D11Texture2D* BackBuffer;
hr = SwapChain->GetBuffer(0, __uuidof(ID3D11Texture2D), (void**)&BackBuffer);
//Create our Render Target
hr = d3d11Device->CreateRenderTargetView(BackBuffer, NULL, &renderTargetView);
BackBuffer->Release();
////////////////////////////////////////////////////////////////////////EXPERIMENT AREA//////////////////////////////////////////////////////////////////////////////////////
ZeroMemory(&textureDesc, sizeof(textureDesc));
// Setup the texture description.
// We will have our map be a square
// We will need to have this texture bound as a render target AND a shader resource
textureDesc.Width = Width/ 3.9729999999999999999999999999999;
textureDesc.Height = Height/3.9729999999999999999999999999999;
textureDesc.MipLevels = 1;
textureDesc.ArraySize = 1;
textureDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
textureDesc.SampleDesc.Count = 1;
textureDesc.SampleDesc.Quality = 0;
textureDesc.Usage = D3D11_USAGE_DEFAULT;
textureDesc.BindFlags = D3D11_BIND_RENDER_TARGET | D3D11_BIND_SHADER_RESOURCE;
textureDesc.CPUAccessFlags = 0;
textureDesc.MiscFlags = 0;
d3d11Device->CreateTexture2D(&textureDesc, NULL, &renderTargetTextureMap);
// Setup the description of the render target view.
renderTargetViewDesc.Format = textureDesc.Format;
renderTargetViewDesc.ViewDimension = D3D11_RTV_DIMENSION_TEXTURE2D;
renderTargetViewDesc.Texture2D.MipSlice = 0;
// Create the render target view.
d3d11Device->CreateRenderTargetView(renderTargetTextureMap, &renderTargetViewDesc, &renderTargetViewMap);
/////////////////////// Map's Shader Resource View
// Setup the description of the shader resource view.
shaderResourceViewDesc.Format = textureDesc.Format;
shaderResourceViewDesc.ViewDimension = D3D11_SRV_DIMENSION_TEXTURE2D;
shaderResourceViewDesc.Texture2D.MostDetailedMip = 0;
shaderResourceViewDesc.Texture2D.MipLevels = 1;
// Create the shader resource view.
d3d11Device->CreateShaderResourceView(renderTargetTextureMap, &shaderResourceViewDesc, &shaderResourceViewMap);
d3d11DevCon->OMSetRenderTargets(1, &renderTargetViewMap, NULL);
d3d11DevCon->PSSetShaderResources(0, 1, &shaderResourceViewMap);
return true;
}
void CleanUp()
{
//Release the COM Objects we created
SwapChain->Release();
d3d11Device->Release();
d3d11DevCon->Release();
renderTargetView->Release();
triangleVertBuffer->Release();
VS->Release();
PS->Release();
VS_Buffer->Release();
PS_Buffer->Release();
vertLayout->Release();
}
bool InitScene()
{
//Compile Shaders from shader file
hr = D3DX11CompileFromFile(L"Effect.fx", 0, 0, "VS", "vs_5_0", 0, 0, 0, &VS_Buffer, 0, 0);
hr = D3DX11CompileFromFile(L"Effect.fx", 0, 0, "PS", "ps_5_0", 0, 0, 0, &PS_Buffer, 0, 0);
//Create the Shader Objects
hr = d3d11Device->CreateVertexShader(VS_Buffer->GetBufferPointer(), VS_Buffer->GetBufferSize(), NULL, &VS);
hr = d3d11Device->CreatePixelShader(PS_Buffer->GetBufferPointer(), PS_Buffer->GetBufferSize(), NULL, &PS);
//Set Vertex and Pixel Shaders
d3d11DevCon->VSSetShader(VS, 0, 0);
d3d11DevCon->PSSetShader(PS, 0, 0);
//Create the Input Layout
hr = d3d11Device->CreateInputLayout(layout, numElements, VS_Buffer->GetBufferPointer(),
VS_Buffer->GetBufferSize(), &vertLayout);
//Set the Input Layout
d3d11DevCon->IASetInputLayout(vertLayout);
//Set Primitive Topology
d3d11DevCon->IASetPrimitiveTopology(D3D11_PRIMITIVE_TOPOLOGY_TRIANGLELIST);
//Create the Viewport
D3D11_VIEWPORT viewport;
ZeroMemory(&viewport, sizeof(D3D11_VIEWPORT));
viewport.TopLeftX = 0;
viewport.TopLeftY = 0;
viewport.Width = 800;
viewport.Height = 600;
//Set the Viewport
d3d11DevCon->RSSetViewports(1, &viewport);
////////////////***********************First Texture Vertex Buffer *******************************/////////////////////////////
//Create the vertex buffer
Vertex v[] =
{
Vertex(-0.5f, -0.5f, 0.0f, 1.0f,0.0f,0.0f, 1.0f),
Vertex(-0.5f, 0.5f, 0.0f, 1.0f,0.0f,0.0f, 1.0f),
Vertex(0.5f, 0.5f, 0.0f, 1.0f, 0.0f,0.0f, 1.0f),
Vertex(0.5f, -0.5f, 0.0f, 1.0f,0.0f, 0.0f, 1.0f),
};
DWORD indices[] = {
// Front Face
0, 1, 3,
1, 2, 3,
};
D3D11_BUFFER_DESC indexBufferDesc;
ZeroMemory(&indexBufferDesc, sizeof(indexBufferDesc));
indexBufferDesc.Usage = D3D11_USAGE_DEFAULT;
indexBufferDesc.ByteWidth = sizeof(DWORD) * 2 * 3;
indexBufferDesc.BindFlags = D3D11_BIND_INDEX_BUFFER;
indexBufferDesc.CPUAccessFlags = 0;
indexBufferDesc.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA iinitData;
iinitData.pSysMem = indices;
d3d11Device->CreateBuffer(&indexBufferDesc, &iinitData, &d2dIndexBuffer);
D3D11_BUFFER_DESC vertexBufferDesc;
ZeroMemory(&vertexBufferDesc, sizeof(vertexBufferDesc));
vertexBufferDesc.Usage = D3D11_USAGE_DEFAULT;
vertexBufferDesc.ByteWidth = sizeof(Vertex) * 4;
vertexBufferDesc.BindFlags = D3D11_BIND_VERTEX_BUFFER;
vertexBufferDesc.CPUAccessFlags = 0;
vertexBufferDesc.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA vertexBufferData;
ZeroMemory(&vertexBufferData, sizeof(vertexBufferData));
vertexBufferData.pSysMem = v;
hr = d3d11Device->CreateBuffer(&vertexBufferDesc, &vertexBufferData, &triangleVertBuffer);
////////////////////////////////////////////////////// Second Vertex.
Vertex v2[] = {
// positions // colors // texture coords
Vertex(1.0f, 1.0f, 0.0f,0.0f,1.0f,0.0f,1.0f), // top right
Vertex(1.0f,0.0f, 0.0f,0.0f,1.0f,0.0f,1.0f), // bottom right
Vertex(0.0f,0.0f, 0.0f,0.0f,1.0f,0.0f,1.0f), // bottom left
Vertex(0.0f, 1.0, 0.0f,0.0f, 1.0f,0.0f,1.0f) // top left
};
DWORD indices2[] = {
// Front Face
0, 1, 2,
0, 2, 3,
};
D3D11_BUFFER_DESC indexBufferDesc2;
ZeroMemory(&indexBufferDesc2, sizeof(indexBufferDesc2));
indexBufferDesc2.Usage = D3D11_USAGE_DEFAULT;
indexBufferDesc2.ByteWidth = sizeof(DWORD) * 2 * 3;
indexBufferDesc2.BindFlags = D3D11_BIND_INDEX_BUFFER;
indexBufferDesc2.CPUAccessFlags = 0;
indexBufferDesc2.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA iinitData2;
iinitData2.pSysMem = indices2;
d3d11Device->CreateBuffer(&indexBufferDesc2, &iinitData2, &d2dIndexBuffer2);
D3D11_BUFFER_DESC vertexBufferDesc2;
ZeroMemory(&vertexBufferDesc2, sizeof(vertexBufferDesc2));
vertexBufferDesc2.Usage = D3D11_USAGE_DEFAULT;
vertexBufferDesc2.ByteWidth = sizeof(Vertex) * 4;
vertexBufferDesc2.BindFlags = D3D11_BIND_VERTEX_BUFFER;
vertexBufferDesc2.CPUAccessFlags = 0;
vertexBufferDesc2.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA vertexBufferData2;
ZeroMemory(&vertexBufferData2, sizeof(vertexBufferData2));
vertexBufferData2.pSysMem = v2;
hr = d3d11Device->CreateBuffer(&vertexBufferDesc2, &vertexBufferData2, &triangleVertBuffer2);
UINT stride = sizeof(Vertex);
UINT offset = 0;
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
return true;
}
Shader File:
struct VS_OUTPUT
{
float4 Pos : SV_POSITION;
float4 Color : COLOR;
};
Texture2D ObjTexture;
SamplerState ObjSamplerState;
VS_OUTPUT VS(float4 inPos : POSITION, float4 inColor : COLOR)
{
VS_OUTPUT output;
output.Pos = inPos;
output.Color = inColor;
return output;
}
float4 PS(VS_OUTPUT input) : SV_TARGET
{
float2 temp;
temp = input.Pos;
float4 diffuse = ObjTexture.Sample(ObjSamplerState,0.5*temp);
return input.Color + diffuse ;
}
return DefWindowProc(hwnd,
msg,
wParam,
lParam);
}
Edit 3:
My vertex Structure:
struct Vertex //Overloaded Vertex Structure
{
Vertex() {}
Vertex(float x, float y, float z,
float cr, float cg, float cb, float ca)
: pos(x, y, z), color(cr, cg, cb, ca) {}
XMFLOAT3 pos;
XMFLOAT4 color;
};
Input description:
D3D11_INPUT_ELEMENT_DESC layout[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D11_INPUT_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D11_INPUT_PER_VERTEX_DATA, 0 },
};
I think so issue is when i pass my texture and read it in diffuse in pixel shader, then mapping of my texture on back buffer is not same and hence i am only finding red color from 2nd scan line and hence resultant zero is produced..?
Edit 3:
My initialization code:
struct Vertex //Overloaded Vertex Structure
{
Vertex() {}
Vertex(float x, float y, float z,
float cr, float cg, float cb, float ca)
: pos(x, y, z), color(cr, cg, cb, ca) {}
XMFLOAT3 pos;
XMFLOAT4 color;
};
ID3D11Texture2D* renderTargetTextureMap;
ID3D11RenderTargetView* renderTargetViewMap;
ID3D11ShaderResourceView* shaderResourceViewMap;
ID3D11SamplerState* CubesTexSamplerState;
ID3D11Buffer *d2dIndexBuffer;
ID3D11Buffer *d2dIndexBuffer2;
D3D11_TEXTURE2D_DESC textureDesc;
D3D11_RENDER_TARGET_VIEW_DESC renderTargetViewDesc;
D3D11_SHADER_RESOURCE_VIEW_DESC shaderResourceViewDesc;
D3D11_INPUT_ELEMENT_DESC layout[] =
{
{ "POSITION", 0, DXGI_FORMAT_R32G32B32_FLOAT, 0, 0, D3D11_INPUT_PER_VERTEX_DATA, 0 },
{ "COLOR", 0, DXGI_FORMAT_R32G32B32A32_FLOAT, 0, 12, D3D11_INPUT_PER_VERTEX_DATA, 0 },
};
UINT numElements = ARRAYSIZE(layout);
bool InitializeDirect3d11App(HINSTANCE hInstance)
{
//Describe our Buffer
DXGI_MODE_DESC bufferDesc;
ZeroMemory(&bufferDesc, sizeof(DXGI_MODE_DESC));
bufferDesc.Width = Width;
bufferDesc.Height = Height;
bufferDesc.RefreshRate.Numerator = 60;
bufferDesc.RefreshRate.Denominator = 1;
bufferDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
bufferDesc.ScanlineOrdering = DXGI_MODE_SCANLINE_ORDER_UNSPECIFIED;
bufferDesc.Scaling = DXGI_MODE_SCALING_UNSPECIFIED;
//Describe our SwapChain
DXGI_SWAP_CHAIN_DESC swapChainDesc;
ZeroMemory(&swapChainDesc, sizeof(DXGI_SWAP_CHAIN_DESC));
swapChainDesc.BufferDesc = bufferDesc;
swapChainDesc.SampleDesc.Count = 1;
swapChainDesc.SampleDesc.Quality = 0;
swapChainDesc.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT;
swapChainDesc.BufferCount = 1;
swapChainDesc.OutputWindow = hwnd;
swapChainDesc.Windowed = TRUE;
swapChainDesc.SwapEffect = DXGI_SWAP_EFFECT_DISCARD;
//Create our SwapChain
hr = D3D11CreateDeviceAndSwapChain(NULL, D3D_DRIVER_TYPE_HARDWARE, NULL, NULL, NULL, NULL,
D3D11_SDK_VERSION, &swapChainDesc, &SwapChain, &d3d11Device, NULL, &d3d11DevCon);
//Create our BackBuffer
ID3D11Texture2D* BackBuffer;
hr = SwapChain->GetBuffer(0, __uuidof(ID3D11Texture2D), (void**)&BackBuffer);
//Create our Render Target
hr = d3d11Device->CreateRenderTargetView(BackBuffer, NULL, &renderTargetView);
BackBuffer->Release();
////////////////////////////////////////////////////////////////////////EXPERIMENT AREA//////////////////////////////////////////////////////////////////////////////////////
ZeroMemory(&textureDesc, sizeof(textureDesc));
// Setup the texture description.
// We will have our map be a square
// We will need to have this texture bound as a render target AND a shader resource
textureDesc.Width = Width/2;
textureDesc.Height = Height/2;
textureDesc.MipLevels = 1;
textureDesc.ArraySize = 1;
textureDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
textureDesc.SampleDesc.Count = 1;
textureDesc.SampleDesc.Quality = 0;
textureDesc.Usage = D3D11_USAGE_DEFAULT;
textureDesc.BindFlags = D3D11_BIND_RENDER_TARGET | D3D11_BIND_SHADER_RESOURCE;
textureDesc.CPUAccessFlags = 0;
textureDesc.MiscFlags = 0;
d3d11Device->CreateTexture2D(&textureDesc, NULL, &renderTargetTextureMap);
// Setup the description of the render target view.
renderTargetViewDesc.Format = textureDesc.Format;
renderTargetViewDesc.ViewDimension = D3D11_RTV_DIMENSION_TEXTURE2D;
renderTargetViewDesc.Texture2D.MipSlice = 0;
// Create the render target view.
d3d11Device->CreateRenderTargetView(renderTargetTextureMap, &renderTargetViewDesc, &renderTargetView);
/////////////////////// Map's Shader Resource View
// Setup the description of the shader resource view.
shaderResourceViewDesc.Format = textureDesc.Format;
shaderResourceViewDesc.ViewDimension = D3D11_SRV_DIMENSION_TEXTURE2D;
shaderResourceViewDesc.Texture2D.MostDetailedMip = 0;
shaderResourceViewDesc.Texture2D.MipLevels = 1;
// Create the shader resource view.
d3d11Device->CreateShaderResourceView(renderTargetTextureMap, &shaderResourceViewDesc, &shaderResourceViewMap);
//d3d11DevCon->OMSetRenderTargets(1, &renderTargetViewMap, NULL);
//d3d11DevCon->PSSetShaderResources(0, 1, &shaderResourceViewMap);
return true;
}
void CleanUp()
{
//Release the COM Objects we created
SwapChain->Release();
d3d11Device->Release();
d3d11DevCon->Release();
renderTargetView->Release();
triangleVertBuffer->Release();
VS->Release();
PS->Release();
VS_Buffer->Release();
PS_Buffer->Release();
vertLayout->Release();
}
bool InitScene()
{
//Compile Shaders from shader file
hr = D3DX11CompileFromFile(L"Effect.fx", 0, 0, "VS", "vs_5_0", 0, 0, 0, &VS_Buffer, 0, 0);
hr = D3DX11CompileFromFile(L"Effect.fx", 0, 0, "PS", "ps_5_0", 0, 0, 0, &PS_Buffer, 0, 0);
//Create the Shader Objects
hr = d3d11Device->CreateVertexShader(VS_Buffer->GetBufferPointer(), VS_Buffer->GetBufferSize(), NULL, &VS);
hr = d3d11Device->CreatePixelShader(PS_Buffer->GetBufferPointer(), PS_Buffer->GetBufferSize(), NULL, &PS);
//Set Vertex and Pixel Shaders
d3d11DevCon->VSSetShader(VS, 0, 0);
d3d11DevCon->PSSetShader(PS, 0, 0);
//Create the Input Layout
hr = d3d11Device->CreateInputLayout(layout, numElements, VS_Buffer->GetBufferPointer(),
VS_Buffer->GetBufferSize(), &vertLayout);
//Set the Input Layout
d3d11DevCon->IASetInputLayout(vertLayout);
//Set Primitive Topology
d3d11DevCon->IASetPrimitiveTopology(D3D11_PRIMITIVE_TOPOLOGY_TRIANGLELIST);
//Create the Viewport
D3D11_VIEWPORT viewport;
ZeroMemory(&viewport, sizeof(D3D11_VIEWPORT));
viewport.TopLeftX = 0;
viewport.TopLeftY = 0;
viewport.Width = 800;
viewport.Height = 600;
//Set the Viewport
d3d11DevCon->RSSetViewports(1, &viewport);
////////////////***********************First Texture Vertex Buffer *******************************/////////////////////////////
//Create the vertex buffer
Vertex v[] =
{
Vertex(-0.35f, -0.35f, 0.0f, 1.0f,0.0f,0.0f, 1.0f),
Vertex(-0.35f, 0.35f, 0.0f, 1.0f,0.0f,0.0f, 1.0f),
Vertex(0.35f, 0.35f, 0.0f, 1.0f, 0.0f,0.0f, 1.0f),
Vertex(0.35f, -0.35f, 0.0f, 1.0f,0.0f, 0.0f, 1.0f),
};
DWORD indices[] = {
// Front Face
0, 1, 3,
1, 2, 3,
};
D3D11_BUFFER_DESC indexBufferDesc;
ZeroMemory(&indexBufferDesc, sizeof(indexBufferDesc));
indexBufferDesc.Usage = D3D11_USAGE_DEFAULT;
indexBufferDesc.ByteWidth = sizeof(DWORD) * 2 * 3;
indexBufferDesc.BindFlags = D3D11_BIND_INDEX_BUFFER;
indexBufferDesc.CPUAccessFlags = 0;
indexBufferDesc.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA iinitData;
iinitData.pSysMem = indices;
d3d11Device->CreateBuffer(&indexBufferDesc, &iinitData, &d2dIndexBuffer);
D3D11_BUFFER_DESC vertexBufferDesc;
ZeroMemory(&vertexBufferDesc, sizeof(vertexBufferDesc));
vertexBufferDesc.Usage = D3D11_USAGE_DEFAULT;
vertexBufferDesc.ByteWidth = sizeof(Vertex) * 4;
vertexBufferDesc.BindFlags = D3D11_BIND_VERTEX_BUFFER;
vertexBufferDesc.CPUAccessFlags = 0;
vertexBufferDesc.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA vertexBufferData;
ZeroMemory(&vertexBufferData, sizeof(vertexBufferData));
vertexBufferData.pSysMem = v;
hr = d3d11Device->CreateBuffer(&vertexBufferDesc, &vertexBufferData, &triangleVertBuffer);
////////////////////////////////////////////////////// Second Vertex.
Vertex v2[] = {
// positions // colors // texture coords
Vertex(1.0f, 1.0f, 0.0f,0.0f,1.0f,0.0f,1.0f), // top right
Vertex(1.0f,0.0f, 0.0f,0.0f,1.0f,0.0f,1.0f), // bottom right
Vertex(0.0f,0.0f, 0.0f,0.0f,1.0f,0.0f,1.0f), // bottom left
Vertex(0.0f, 1.0, 0.0f,0.0f, 1.0f,0.0f,1.0f) // top left
};
DWORD indices2[] = {
// Front Face
0, 1, 2,
0, 2, 3,
};
D3D11_BUFFER_DESC indexBufferDesc2;
ZeroMemory(&indexBufferDesc2, sizeof(indexBufferDesc2));
indexBufferDesc2.Usage = D3D11_USAGE_DEFAULT;
indexBufferDesc2.ByteWidth = sizeof(DWORD) * 2 * 3;
indexBufferDesc2.BindFlags = D3D11_BIND_INDEX_BUFFER;
indexBufferDesc2.CPUAccessFlags = 0;
indexBufferDesc2.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA iinitData2;
iinitData2.pSysMem = indices2;
d3d11Device->CreateBuffer(&indexBufferDesc2, &iinitData2, &d2dIndexBuffer2);
D3D11_BUFFER_DESC vertexBufferDesc2;
ZeroMemory(&vertexBufferDesc2, sizeof(vertexBufferDesc2));
vertexBufferDesc2.Usage = D3D11_USAGE_DEFAULT;
vertexBufferDesc2.ByteWidth = sizeof(Vertex) * 4;
vertexBufferDesc2.BindFlags = D3D11_BIND_VERTEX_BUFFER;
vertexBufferDesc2.CPUAccessFlags = 0;
vertexBufferDesc2.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA vertexBufferData2;
ZeroMemory(&vertexBufferData2, sizeof(vertexBufferData2));
vertexBufferData2.pSysMem = v2;
hr = d3d11Device->CreateBuffer(&vertexBufferDesc2, &vertexBufferData2, &triangleVertBuffer2);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
return true;
}
Edit 4:
Added draw scene code:
void DrawScene()
{
//Set the vertex buffer
UINT stride = sizeof(Vertex);
UINT offset = 0;
///////////////////////////////////////////////////////////Buffer 2//////////////////////////////////////////////////////////////////
d3d11DevCon->IASetIndexBuffer(d2dIndexBuffer, DXGI_FORMAT_R32_UINT, 0);
d3d11DevCon->IASetVertexBuffers(0, 1, &triangleVertBuffer, &stride, &offset);
////Draw the triangle
d3d11DevCon->DrawIndexed(6, 0, 0);
d3d11DevCon->OMSetRenderTargets(1, &renderTargetView, NULL);
////////////////////////////////////////////
d3d11DevCon->PSSetShaderResources(0, 1, &shaderResourceViewMap);
d3d11DevCon->IASetIndexBuffer(d2dIndexBuffer2, DXGI_FORMAT_R32_UINT, 0);
d3d11DevCon->IASetVertexBuffers(0, 1, &triangleVertBuffer2, &stride, &offset);
////Draw the triangle
d3d11DevCon->DrawIndexed(6, 0, 0);
//Present the backbuffer to the screen
SwapChain->Present(0, 0);
}
In order to calculate correct texture coordinates from the vertex Position, you need to:
1.Create the render target texture with Width and Height (not the halves):
textureDesc.Width = Width; // /2; // Do not use half width
textureDesc.Height = Height; // /2; // Do not use half height
2.Divide the position to a float2(1 / Width, 1 / Height) in your pixel shader, like this:
float2 tex = input.Pos * float2(1.0f / 800.0f, 1.0f / 600.0f);
float4 diffuse = ObjTexture.Sample(ObjSamplerState, tex);
return input.Color + diffuse;
I am trying to read some data from a texture2d in DirectX11 compute shader, however, the 'Load' function of a texture2D object keeps returning 0 even though the texture object is filled with the same float number.
This is a 160 * 120 texture2d with DXGI_FORMAT_R32G32B32A32_FLOAT. The following code is how I created this resource:
HRESULT TestResources(ID3D11Device* pd3dDevice, ID3D11DeviceContext* pImmediateContext) {
float *test = new float[4 * 80 * 60 * 4]; // 80 * 60, 4 channels, 1 big texture contains 4 80 * 60 subimage
for (int i = 0; i < 4 * 80 * 60 * 4; i++) test[i] = 0.7f;
HRESULT hr = S_OK;
D3D11_TEXTURE2D_DESC RTtextureDesc;
ZeroMemory(&RTtextureDesc, sizeof(D3D11_TEXTURE2D_DESC));
RTtextureDesc.Width = 160;
RTtextureDesc.Height = 120;
RTtextureDesc.MipLevels = 1;
RTtextureDesc.ArraySize = 1;
RTtextureDesc.Format = DXGI_FORMAT_R32G32B32A32_FLOAT;
RTtextureDesc.SampleDesc.Count = 1;
RTtextureDesc.SampleDesc.Quality = 0;
RTtextureDesc.Usage = D3D11_USAGE_DYNAMIC;
RTtextureDesc.BindFlags = D3D11_BIND_SHADER_RESOURCE;
RTtextureDesc.CPUAccessFlags = D3D11_CPU_ACCESS_WRITE;
RTtextureDesc.MiscFlags = 0;
D3D11_SUBRESOURCE_DATA InitData;
InitData.pSysMem = test;
InitData.SysMemPitch = sizeof(float) * 4;
V_RETURN(pd3dDevice->CreateTexture2D(&RTtextureDesc, &InitData, &m_pInputTex2Ds));
//V_RETURN(pd3dDevice->CreateTexture2D(&RTtextureDesc, NULL, &m_pInputTex2Ds));
D3D11_SHADER_RESOURCE_VIEW_DESC SRViewDesc;
ZeroMemory(&SRViewDesc, sizeof(SRViewDesc));
SRViewDesc.Format = RTtextureDesc.Format;
SRViewDesc.ViewDimension = D3D11_SRV_DIMENSION_TEXTURE2D;
SRViewDesc.Texture2D.MostDetailedMip = 0;
SRViewDesc.Texture2D.MipLevels = 1;
V_RETURN(pd3dDevice->CreateShaderResourceView(m_pInputTex2Ds, &SRViewDesc, &m_pInputTexSRV));
delete[] test;
return hr;
}
And then I try to run dispatch with X = Y = 2 and Z = 1 like the following:
void ComputeShaderReduction::ExecuteComputeShader(ID3D11DeviceContext* pd3dImmediateContext, UINT uInputNum, ID3D11UnorderedAccessView** ppUAVInputs, UINT X, UINT Y, UINT Z) {
pd3dImmediateContext->CSSetShader(m_pComputeShader, nullptr, 0);
pd3dImmediateContext->CSSetShaderResources(0, 1, &m_pInputTexSRV); // test code
pd3dImmediateContext->CSSetUnorderedAccessViews(0, uInputNum, ppUAVInputs, nullptr);
//pd3dImmediateContext->CSSetUnorderedAccessViews(0, 1, &m_pGPUOutUAVs, nullptr);
pd3dImmediateContext->UpdateSubresource(m_pConstBuf, 0, nullptr, &m_ConstBuf, 0, 0);
pd3dImmediateContext->CSSetConstantBuffers(0, 1, &m_pConstBuf);
pd3dImmediateContext->Dispatch(X, Y, Z);
pd3dImmediateContext->CSSetShader(nullptr, nullptr, 0);
ID3D11UnorderedAccessView* ppUAViewnullptr[1] = { nullptr };
pd3dImmediateContext->CSSetUnorderedAccessViews(0, 1, ppUAViewnullptr, nullptr);
ID3D11ShaderResourceView* ppSRVnullptr[1] = { nullptr };
pd3dImmediateContext->CSSetShaderResources(0, 1, ppSRVnullptr);
ID3D11Buffer* ppCBnullptr[1] = { nullptr };
pd3dImmediateContext->CSSetConstantBuffers(0, 1, ppCBnullptr);
}
And I wrote a very simple CS shader to try to get the data in the texture2d and out it. So, the compute shader looks like this:
#define subimg_dim_x 80
#define subimg_dim_y 60
Texture2D<float4> BufferIn : register(t0);
StructuredBuffer<float> Test: register(t1);
RWStructuredBuffer<float> BufferOut : register(u0);
groupshared float sdata[subimg_dim_x];
[numthreads(subimg_dim_x, 1, 1)]
void CSMain(uint3 DTid : SV_DispatchThreadID,
uint3 threadIdx : SV_GroupThreadID,
uint3 groupIdx : SV_GroupID) {
sdata[threadIdx.x] = 0.0;
GroupMemoryBarrierWithGroupSync();
if (threadIdx.x == 0) {
float4 num = BufferIn.Load(uint3(groupIdx.x, groupIdx.y, 1));
//BufferOut[groupIdx.y * 2 + groupIdx.x] = 2.0; //This one gives me 2.0 as output in the console
BufferOut[groupIdx.y * 2 + groupIdx.x] = num.x; //This one keeps giving me 0.0 and in the texture, r = g = b = a = 0.7 or x = y = z = w = 0.7, so it suppose to print 0.7 in the console.
}
GroupMemoryBarrierWithGroupSync();
}
I think the way I print the CS shader result on CPU end is correct.
void ComputeShaderReduction::CopyToCPUBuffer(ID3D11Device* pdevice, ID3D11DeviceContext* pd3dImmediateContext, ID3D11Buffer* pGPUOutBufs) {
D3D11_BUFFER_DESC desc;
ZeroMemory(&desc, sizeof(desc));
pGPUOutBufs->GetDesc(&desc);
desc.CPUAccessFlags = D3D11_CPU_ACCESS_READ;
desc.Usage = D3D11_USAGE_STAGING;
desc.BindFlags = 0;
desc.MiscFlags = 0;
if (!m_pCPUOutBufs && SUCCEEDED(pdevice->CreateBuffer(&desc, nullptr, &m_pCPUOutBufs))) {
pd3dImmediateContext->CopyResource(m_pCPUOutBufs, pGPUOutBufs);
}
else pd3dImmediateContext->CopyResource(m_pCPUOutBufs, pGPUOutBufs);
D3D11_MAPPED_SUBRESOURCE MappedResource;
float *p;
pd3dImmediateContext->Map(m_pCPUOutBufs, 0, D3D11_MAP_READ, 0, &MappedResource);
p = (float*)MappedResource.pData;
for (int i = 0; i < 4; i++) printf("%d %f\n", i, p[i]);
pd3dImmediateContext->Unmap(m_pCPUOutBufs, 0);
printf("\n");
}
The buffer that bind to UAV has only 4 elements. So, if all the float numbers in my texture2d are 0.7, I should have 4 0.7s get printed in CopyToCPUBuffer function instead of 0.0s.
Is anyone know what could be wrong in my code or can someone provide me an entire example or a tutorial that shows how to read DirectX 11 texture2d's data in compute shader correctly?
Thanks in advance.
The following is wrong for a start. The Pitch of your input data is the number of bytes per row of the texture, not per pixel.
InitData.SysMemPitch = sizeof(float) * 4;
Secondly:
float4 num = BufferIn.Load(uint3(groupIdx.x, groupIdx.y, 1));
You're trying to load data from the 2nd mip of the texture, it only has 1 mip level.
I am trying to implement online face recognition using the webcam. I am using this two websites as references
shervinemami.co.cc
cognotics.com
I have few questions:
In face recognition, there are 6 steps:
Grab a frame from the camera
Detect a face within the image
Crop the frame to show just the face
Convert the frame to greyscale
Preprocess the image
Recognize the person in the image.
I am able to do the first five steps. Last step i am not able to do. I am not sure how to link step 5 to step 6.
I have already created the train.txt file and test.txt file which contains the information of the training and testing images. I have already added the functions such as learn(), doPCA() to the code...
But the point is how to use these functions in the main to recognize the image that is already preprocessed.
Need some help on it...
Attached the code below:
// Real-time.cpp : Defines the entry point for the console application.
#include "stdafx.h"
#include <cv.h>
#include <cxcore.h>
#include <highgui.h>
#include <cvaux.h>
IplImage ** faceImgArr = 0; // array of face images
CvMat * personNumTruthMat = 0; // array of person numbers
int nTrainFaces = 0; // the number of training images
int nEigens = 0; // the number of eigenvalues
IplImage * pAvgTrainImg = 0; // the average image
IplImage ** eigenVectArr = 0; // eigenvectors
CvMat * eigenValMat = 0; // eigenvalues
CvMat * projectedTrainFaceMat = 0; // projected training faces
IplImage* getCameraFrame(CvCapture* &camera);
IplImage* detectFaces( IplImage *img ,CvHaarClassifierCascade* facecascade,CvMemStorage* storage );
CvRect detectFaceInImage(IplImage *inputImg, CvHaarClassifierCascade* cascade);
IplImage* preprocess( IplImage* inputImg);
IplImage* resizeImage(const IplImage *origImg, int newWidth,
int newHeight, bool keepAspectRatio);
void learn();
void recognize();
void doPCA();
void storeTrainingData();
int loadTrainingData(CvMat ** pTrainPersonNumMat);
int findNearestNeighbor(float * projectedTestFace);
int loadFaceImgArray(char * filename);
int _tmain(int argc, _TCHAR* argv[])
{
CvCapture* camera = 0; // The camera device.
CvMemStorage *storage;
cvNamedWindow( "Realtime:", CV_WINDOW_AUTOSIZE);
char *faceCascadeFilename = "C:/OpenCV2.1/data/haarcascades/haarcascade_frontalface_alt.xml";
CvHaarClassifierCascade* faceCascade;
faceCascade = (CvHaarClassifierCascade*)cvLoad(faceCascadeFilename, 0, 0, 0);
storage = cvCreateMemStorage( 0 );
learn();
while ( cvWaitKey(10) != 27 ) // Quit on "Escape" key
{
IplImage *frame = getCameraFrame(camera);
//IplImage* resized=cvCreateImage(cvSize(420,240),frame->depth,3);
//cvResizeWindow( "Image:", 640, 480);
//cvResize(frame,resized);
//cvShowImage( "Realtime:", resized );
IplImage *imgA = resizeImage(frame, 420,240, true);
IplImage *frame1 = detectFaces(imgA,faceCascade,storage);
frame1 = preprocess(frame1);
}
// Free the camera.
cvReleaseCapture( &camera );
cvReleaseMemStorage( &storage );
return 0;
}
IplImage* getCameraFrame(CvCapture* &camera)
{
IplImage *frame;
int w, h;
// If the camera hasn't been initialized, then open it.
if (!camera) {
printf("Acessing the camera ...\n");
camera = cvCreateCameraCapture( 0 );
if (!camera) {
printf("Couldn't access the camera.\n");
exit(1);
}
// Try to set the camera resolution to 320 x 240.
cvSetCaptureProperty(camera, CV_CAP_PROP_FRAME_WIDTH, 320);
cvSetCaptureProperty(camera, CV_CAP_PROP_FRAME_HEIGHT, 240);
// Get the first frame, to make sure the camera is initialized.
frame = cvQueryFrame( camera );
if (frame) {
w = frame->width;
h = frame->height;
printf("Got the camera at %dx%d resolution.\n", w, h);
}
// Wait a little, so that the camera can auto-adjust its brightness.
Sleep(1000); // (in milliseconds)
}
// Wait until the next camera frame is ready, then grab it.
frame = cvQueryFrame( camera );
if (!frame) {
printf("Couldn't grab a camera frame.\n");
exit(1);
}
return frame;
}
CvRect detectFaceInImage(IplImage *inputImg, CvHaarClassifierCascade* cascade)
{
// Smallest face size.
CvSize minFeatureSize = cvSize(20, 20);
// Only search for 1 face.
int flags = CV_HAAR_FIND_BIGGEST_OBJECT | CV_HAAR_DO_ROUGH_SEARCH;
// How detailed should the search be.
float search_scale_factor = 1.1f;
IplImage *detectImg;
IplImage *greyImg = 0;
CvMemStorage* storage;
CvRect rc;
double t;
CvSeq* rects;
CvSize size;
int i, ms, nFaces;
storage = cvCreateMemStorage(0);
cvClearMemStorage( storage );
// If the image is color, use a greyscale copy of the image.
detectImg = (IplImage*)inputImg;
if (inputImg->nChannels > 1) {
size = cvSize(inputImg->width, inputImg->height);
greyImg = cvCreateImage(size, IPL_DEPTH_8U, 1 );
cvCvtColor( inputImg, greyImg, CV_BGR2GRAY );
detectImg = greyImg; // Use the greyscale image.
}
// Detect all the faces in the greyscale image.
t = (double)cvGetTickCount();
rects = cvHaarDetectObjects( detectImg, cascade, storage,
search_scale_factor, 3, flags, minFeatureSize);
t = (double)cvGetTickCount() - t;
ms = cvRound( t / ((double)cvGetTickFrequency() * 1000.0) );
nFaces = rects->total;
printf("Face Detection took %d ms and found %d objects\n", ms, nFaces);
// Get the first detected face (the biggest).
if (nFaces > 0)
rc = *(CvRect*)cvGetSeqElem( rects, 0 );
else
rc = cvRect(-1,-1,-1,-1); // Couldn't find the face.
if (greyImg)
cvReleaseImage( &greyImg );
cvReleaseMemStorage( &storage );
//cvReleaseHaarClassifierCascade( &cascade );
return rc; // Return the biggest face found, or (-1,-1,-1,-1).
}
IplImage* detectFaces( IplImage *img ,CvHaarClassifierCascade* facecascade,CvMemStorage* storage )
{
int i;
CvRect *r;
CvSeq *faces = cvHaarDetectObjects(
img,
facecascade,
storage,
1.1,
3,
0 /*CV_HAAR_DO_CANNY_PRUNNING*/,
cvSize( 40, 40 ) );
int padding_width = 30; // pixels
int padding_height = 30; // pixels
for( i = 0 ; i < ( faces ? faces->total : 0 ) ; i++ ) {
r = ( CvRect* )cvGetSeqElem( faces, i );
cvRectangle( img,
cvPoint( r->x, r->y ),
cvPoint( r->x + r->width, r->y + r->height ),
CV_RGB( 255, 0, 0 ), 1, 8, 0 );
}
cvShowImage( "Realtime:", img );
//cropping the face
cvSetImageROI(img, cvRect(r->x,r->y,r->width,r->height));
IplImage *img2 = cvCreateImage(cvGetSize(img),
img->depth,
img->nChannels);
cvCopy(img, img2, NULL);
cvResetImageROI(img);
return img;
}
IplImage* preprocess( IplImage* inputImg){
IplImage *detectImg, *greyImg = 0;
IplImage *imageProcessed;
CvSize size;
detectImg = (IplImage*)inputImg;
if (inputImg->nChannels > 1) {
size = cvSize(inputImg->width, inputImg->height);
greyImg = cvCreateImage(size, IPL_DEPTH_8U, 1 );
cvCvtColor( inputImg, greyImg, CV_BGR2GRAY );
detectImg = greyImg; // Use the greyscale image.
}
imageProcessed = cvCreateImage(cvSize(inputImg->width, inputImg->height), IPL_DEPTH_8U, 1);
cvResize(detectImg, imageProcessed, CV_INTER_LINEAR);
cvEqualizeHist(imageProcessed, imageProcessed);
return imageProcessed;
}
IplImage* resizeImage(const IplImage *origImg, int newWidth,
int newHeight, bool keepAspectRatio)
{
IplImage *outImg = 0;
int origWidth;
int origHeight;
if (origImg) {
origWidth = origImg->width;
origHeight = origImg->height;
}
if (newWidth <= 0 || newHeight <= 0 || origImg == 0
|| origWidth <= 0 || origHeight <= 0) {
//cerr << "ERROR: Bad desired image size of " << newWidth
// << "x" << newHeight << " in resizeImage().\n";
exit(1);
}
if (keepAspectRatio) {
// Resize the image without changing its aspect ratio,
// by cropping off the edges and enlarging the middle section.
CvRect r;
// input aspect ratio
float origAspect = (origWidth / (float)origHeight);
// output aspect ratio
float newAspect = (newWidth / (float)newHeight);
// crop width to be origHeight * newAspect
if (origAspect > newAspect) {
int tw = (origHeight * newWidth) / newHeight;
r = cvRect((origWidth - tw)/2, 0, tw, origHeight);
}
else { // crop height to be origWidth / newAspect
int th = (origWidth * newHeight) / newWidth;
r = cvRect(0, (origHeight - th)/2, origWidth, th);
}
IplImage *croppedImg = cropImage(origImg, r);
// Call this function again, with the new aspect ratio image.
// Will do a scaled image resize with the correct aspect ratio.
outImg = resizeImage(croppedImg, newWidth, newHeight, false);
cvReleaseImage( &croppedImg );
}
else {
// Scale the image to the new dimensions,
// even if the aspect ratio will be changed.
outImg = cvCreateImage(cvSize(newWidth, newHeight),
origImg->depth, origImg->nChannels);
if (newWidth > origImg->width && newHeight > origImg->height) {
// Make the image larger
cvResetImageROI((IplImage*)origImg);
// CV_INTER_LINEAR: good at enlarging.
// CV_INTER_CUBIC: good at enlarging.
cvResize(origImg, outImg, CV_INTER_LINEAR);
}
else {
// Make the image smaller
cvResetImageROI((IplImage*)origImg);
// CV_INTER_AREA: good at shrinking (decimation) only.
cvResize(origImg, outImg, CV_INTER_AREA);
}
}
return outImg;
}
void learn()
{
int i, offset;
// load training data
nTrainFaces = loadFaceImgArray("C:/Users/HP/Desktop/OpenCV/50_images_of_15_people.txt");
if( nTrainFaces < 2 )
{
fprintf(stderr,
"Need 2 or more training faces\n"
"Input file contains only %d\n", nTrainFaces);
return;
}
// do PCA on the training faces
doPCA();
// project the training images onto the PCA subspace
projectedTrainFaceMat = cvCreateMat( nTrainFaces, nEigens, CV_32FC1 );
offset = projectedTrainFaceMat->step / sizeof(float);
for(i=0; i<nTrainFaces; i++)
{
//int offset = i * nEigens;
cvEigenDecomposite(
faceImgArr[i],
nEigens,
eigenVectArr,
0, 0,
pAvgTrainImg,
//projectedTrainFaceMat->data.fl + i*nEigens);
projectedTrainFaceMat->data.fl + i*offset);
}
// store the recognition data as an xml file
storeTrainingData();
}
void recognize()
{
int i, nTestFaces = 0; // the number of test images
CvMat * trainPersonNumMat = 0; // the person numbers during training
float * projectedTestFace = 0;
// load test images and ground truth for person number
nTestFaces = loadFaceImgArray("C:/Users/HP/Desktop/OpenCV/test.txt");
printf("%d test faces loaded\n", nTestFaces);
// load the saved training data
if( !loadTrainingData( &trainPersonNumMat ) ) return;
// project the test images onto the PCA subspace
projectedTestFace = (float *)cvAlloc( nEigens*sizeof(float) );
for(i=0; i<nTestFaces; i++)
{
int iNearest, nearest, truth;
// project the test image onto the PCA subspace
cvEigenDecomposite(
faceImgArr[i],
nEigens,
eigenVectArr,
0, 0,
pAvgTrainImg,
projectedTestFace);
iNearest = findNearestNeighbor(projectedTestFace);
truth = personNumTruthMat->data.i[i];
nearest = trainPersonNumMat->data.i[iNearest];
printf("nearest = %d, Truth = %d\n", nearest, truth);
}
}
int loadTrainingData(CvMat ** pTrainPersonNumMat)
{
CvFileStorage * fileStorage;
int i;
// create a file-storage interface
fileStorage = cvOpenFileStorage( "facedata.xml", 0, CV_STORAGE_READ );
if( !fileStorage )
{
fprintf(stderr, "Can't open facedata.xml\n");
return 0;
}
nEigens = cvReadIntByName(fileStorage, 0, "nEigens", 0);
nTrainFaces = cvReadIntByName(fileStorage, 0, "nTrainFaces", 0);
*pTrainPersonNumMat = (CvMat *)cvReadByName(fileStorage, 0, "trainPersonNumMat", 0);
eigenValMat = (CvMat *)cvReadByName(fileStorage, 0, "eigenValMat", 0);
projectedTrainFaceMat = (CvMat *)cvReadByName(fileStorage, 0, "projectedTrainFaceMat", 0);
pAvgTrainImg = (IplImage *)cvReadByName(fileStorage, 0, "avgTrainImg", 0);
eigenVectArr = (IplImage **)cvAlloc(nTrainFaces*sizeof(IplImage *));
for(i=0; i<nEigens; i++)
{
char varname[200];
sprintf( varname, "eigenVect_%d", i );
eigenVectArr[i] = (IplImage *)cvReadByName(fileStorage, 0, varname, 0);
}
// release the file-storage interface
cvReleaseFileStorage( &fileStorage );
return 1;
}
void storeTrainingData()
{
CvFileStorage * fileStorage;
int i;
// create a file-storage interface
fileStorage = cvOpenFileStorage( "facedata.xml", 0, CV_STORAGE_WRITE );
// store all the data
cvWriteInt( fileStorage, "nEigens", nEigens );
cvWriteInt( fileStorage, "nTrainFaces", nTrainFaces );
cvWrite(fileStorage, "trainPersonNumMat", personNumTruthMat, cvAttrList(0,0));
cvWrite(fileStorage, "eigenValMat", eigenValMat, cvAttrList(0,0));
cvWrite(fileStorage, "projectedTrainFaceMat", projectedTrainFaceMat, cvAttrList(0,0));
cvWrite(fileStorage, "avgTrainImg", pAvgTrainImg, cvAttrList(0,0));
for(i=0; i<nEigens; i++)
{
char varname[200];
sprintf( varname, "eigenVect_%d", i );
cvWrite(fileStorage, varname, eigenVectArr[i], cvAttrList(0,0));
}
// release the file-storage interface
cvReleaseFileStorage( &fileStorage );
}
int findNearestNeighbor(float * projectedTestFace)
{
//double leastDistSq = 1e12;
double leastDistSq = DBL_MAX;
int i, iTrain, iNearest = 0;
for(iTrain=0; iTrain<nTrainFaces; iTrain++)
{
double distSq=0;
for(i=0; i<nEigens; i++)
{
float d_i =
projectedTestFace[i] -
projectedTrainFaceMat->data.fl[iTrain*nEigens + i];
//distSq += d_i*d_i / eigenValMat->data.fl[i]; // Mahalanobis
distSq += d_i*d_i; // Euclidean
}
if(distSq < leastDistSq)
{
leastDistSq = distSq;
iNearest = iTrain;
}
}
return iNearest;
}
void doPCA()
{
int i;
CvTermCriteria calcLimit;
CvSize faceImgSize;
// set the number of eigenvalues to use
nEigens = nTrainFaces-1;
// allocate the eigenvector images
faceImgSize.width = faceImgArr[0]->width;
faceImgSize.height = faceImgArr[0]->height;
eigenVectArr = (IplImage**)cvAlloc(sizeof(IplImage*) * nEigens);
for(i=0; i<nEigens; i++)
eigenVectArr[i] = cvCreateImage(faceImgSize, IPL_DEPTH_32F, 1);
// allocate the eigenvalue array
eigenValMat = cvCreateMat( 1, nEigens, CV_32FC1 );
// allocate the averaged image
pAvgTrainImg = cvCreateImage(faceImgSize, IPL_DEPTH_32F, 1);
// set the PCA termination criterion
calcLimit = cvTermCriteria( CV_TERMCRIT_ITER, nEigens, 1);
// compute average image, eigenvalues, and eigenvectors
cvCalcEigenObjects(
nTrainFaces,
(void*)faceImgArr,
(void*)eigenVectArr,
CV_EIGOBJ_NO_CALLBACK,
0,
0,
&calcLimit,
pAvgTrainImg,
eigenValMat->data.fl);
cvNormalize(eigenValMat, eigenValMat, 1, 0, CV_L1, 0);
}
int loadFaceImgArray(char * filename)
{
FILE * imgListFile = 0;
char imgFilename[512];
int iFace, nFaces=0;
// open the input file
if( !(imgListFile = fopen(filename, "r")) )
{
fprintf(stderr, "Can\'t open file %s\n", filename);
return 0;
}
// count the number of faces
while( fgets(imgFilename, 512, imgListFile) ) ++nFaces;
rewind(imgListFile);
// allocate the face-image array and person number matrix
faceImgArr = (IplImage **)cvAlloc( nFaces*sizeof(IplImage *) );
personNumTruthMat = cvCreateMat( 1, nFaces, CV_32SC1 );
// store the face images in an array
for(iFace=0; iFace<nFaces; iFace++)
{
// read person number and name of image file
fscanf(imgListFile,
"%d %s", personNumTruthMat->data.i+iFace, imgFilename);
// load the face image
faceImgArr[iFace] = cvLoadImage(imgFilename, CV_LOAD_IMAGE_GRAYSCALE);
if( !faceImgArr[iFace] )
{
fprintf(stderr, "Can\'t load image from %s\n", imgFilename);
return 0;
}
}
fclose(imgListFile);
return nFaces;
}
My answer may came late but it might be useful for pals if i answer it.I am working on a similar project and i have faced the same problem.I solved it by writing a function the saves or write the detected,cropped and preprocessed image on to the hard disk of my computer(Using CvWrite).And feeding the parameter of the saved images to the recognition part of the code. It has made my life easier.It has been a bit harder for me to to pass the parameters of the rect of the region of interest. If you or someone else did this it might be great sharing the code with us.
You can use the following code to save the image after resizing it to a constant value using the resizeimage function on you code.
void saveCroppedFaces(CvSeq* tempon,IplImage* DetectedImage)
{
char* name;
int nFaces;
CvRect rect;
nFaces=tempon->total;
name =new char[nFaces];
IplImage* cropped = 0;
IplImage* croppedResized=0;
Mat croped;
for(int k=0;k<nFaces;k++)
{
itoa(k,(name+k),10);
rect = *(CvRect*)cvGetSeqElem( tempon, k );
cropped= cropImage(DetectedImage,rect);
//i can resize the cropped faces in to a fixed size here
//i can write a function to save images and call it so
//that it will save it in to hard drive
//cvNamedWindow((name+k),CV_WINDOW_AUTOSIZE);
//cvShowImage((name+k),cropped);
croppedResized=resizeImage(cropped,60,60);
croped=IplToMatConverter(croppedResized);
saveROI(croped,itoa(k,(name+k),10));
cvReleaseImage(&cropped);
}
name=NULL;
delete[] name;
}
void saveROI(Mat mat,String outputFileName)
{
string store_path("C://Users/sizusuzu/Desktop/Images/FaceDetection2
/"+outputFileName+".jpg");
bool write_success = imwrite(store_path,mat);
}
After this you can change the IplImage* to Mat using
Mat IplToMatConverter(IplImage* imageToMat)
{
Mat mat = cvarrToMat(imageToMat);
return mat;
}
And use the Mat in FaceRecognizer API.Or just do the other/harder way.
Thanks
I just read
int _tmain(int argc, _TCHAR* argv[])
{
.......
}
part of your code. This code is used for detecting the face in the image. Lets say it is Face_x. Now extract features from Face_x, call it as F_x. In your database, you should store features {F_1, F_2,..., F_N} extracted from n different faces {Face_1, Face_2,..Face_N}.
Simple algorithm to recognize Face_x is to calculate Euclidean distances between F_x and n features. The minimum distance (below threshold) gives corresponding face. If the minimum distance is not below threshold then Face_x is a new face. Add feature F_x to database. This way you can increase your database. You can begin your algorithm with no features in database. With each new face, database grows.
I hope the method suggested by me will lead you to the solution