如何将RTV输出复制到cubemap的一侧？

Question

我目前正在实现漫射灌溉(基于图像的闪电PBR)在我的游戏引擎中。我到了一个点，我必须采取一个HDR图像，并把它变成一个立方体地图。我目前使用的EquirectangularToCubemap着色器和它的工作罚款。我能够将HDR映像投影到一个(单位)立方体上。现在是我被卡住的部分，我不能把这个立方体变成一个cubemap。我尝试使用ShaderResourceView.和1 TextureCube，6 RenderTargetView's，我的计划是用90的FOV从不同的视图投影渲染6次(单元)立方体，以捕捉每个渲染目标中的整个边，最后将渲染目标的每个输出复制到纹理立方体的对应侧。我不知道如何做到这一点。

注意：--我使用DirextX11作为呈现后端。下面是关于我的问题的伪代码(不起作用)

   glm::mat4 captureProjection = glm::perspective(glm::radians(90.0f), 1.0f, 0.1f, 10.0f);
   glm::mat4 captureViews[] =
    {
        glm::lookAt(glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(1.0f,  0.0f,  0.0f), glm::vec3(0.0f, -1.0f,  0.0f)),
        glm::lookAt(glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(-1.0f, 0.0f,  0.0f), glm::vec3(0.0f, -1.0f,  0.0f)),
        glm::lookAt(glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(0.0f,  1.0f,  0.0f), glm::vec3(0.0f,  0.0f,  1.0f)),
        glm::lookAt(glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(0.0f, -1.0f,  0.0f), glm::vec3(0.0f,  0.0f, -1.0f)),
        glm::lookAt(glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(0.0f,  0.0f,  1.0f), glm::vec3(0.0f, -1.0f,  0.0f)),
        glm::lookAt(glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(0.0f,  0.0f, -1.0f), glm::vec3(0.0f, -1.0f,  0.0f))
    };
 
    //Create the TextureCube
    D3D11_TEXTURE2D_DESC textureDesc = {};
    textureDesc.Width = 512;
    textureDesc.Height = 512;
    textureDesc.MipLevels = 1;
    textureDesc.ArraySize = 6;
    textureDesc.Format = DXGI_FORMAT_R32G32B32A32_FLOAT;
    textureDesc.CPUAccessFlags = 0;
    textureDesc.SampleDesc.Count = 1;
    textureDesc.SampleDesc.Quality = 0;
    textureDesc.Usage = D3D11_USAGE_DEFAULT;
    textureDesc.BindFlags = D3D11_BIND_SHADER_RESOURCE | D3D11_BIND_RENDER_TARGET;
    textureDesc.MiscFlags = D3D11_RESOURCE_MISC_TEXTURECUBE;
    ID3D11Texture2D* tex = nullptr;
    DX_CALL(DX11Internal::GetDevice()->CreateTexture2D(&textureDesc, nullptr, &tex));
    // Create the Shader Resource view for the texture cube
    D3D11_SHADER_RESOURCE_VIEW_DESC srvDesc = {};
    srvDesc.Format = textureDesc.Format;
    srvDesc.ViewDimension = D3D11_SRV_DIMENSION_TEXTURECUBE;
    srvDesc.Texture2D.MostDetailedMip = 0;
    srvDesc.Texture2D.MipLevels = 1;
    DX_CALL(DX11Internal::GetDevice()->CreateShaderResourceView(tex, &srvDesc, &mSRV));
 
 
    //Create the Render target views
    Vector<ID3D11RenderTargetView*> rtvs;
 
    for (Uint i = 0; i < 6; i++)
    {
        D3D11_RENDER_TARGET_VIEW_DESC renderTargetViewDesc = {};
        renderTargetViewDesc.Format = textureDesc.Format;
        renderTargetViewDesc.ViewDimension = D3D11_RTV_DIMENSION_TEXTURE2D;
        renderTargetViewDesc.Texture2D.MipSlice = 0;
 
        ID3D11RenderTargetView* view = nullptr;
        DX11Internal::GetDevice()->CreateRenderTargetView(tex, &renderTargetViewDesc, &view);
        rtvs.push_back(view);
    }
    tex->Release();
 
    auto deviceContext = DX11Internal::GetDeviceContext();
    for (Uint i = 0; i < 6; ++i)
    {
        float clearColor[4] = { 1.0f, 0.1f, 0.1f, 1.0f };
        deviceContext->ClearRenderTargetView(rtvs[i], clearColor);
        Vault::Get<Shader>("EquirectangularToCubemap.hlsl")->Bind();
 
        auto data = captureProjection * captureViews[i];
        cbuffer->Bind();
        cbuffer->SetData(&data);
        texture->Bind(0);
        tempPipeline->Bind();
        deviceContext->OMSetRenderTargets(1, &rtvs[i], nullptr);
//I am rendering the cube here from different view projection to capture the faces, but I dont't know where to copy the data to the //side of the TextureCube :( [Note that I am doing this only once]
        RenderCommand::DrawIndexed(tempPipeline, 36);
    }

提前感谢！

Answer 1

在创建呈现视图时，您没有指定要写入哪些片。切片的正确描述是：

D3D11_RENDER_TARGET_VIEW_DESC renderTargetViewDesc = {};
renderTargetViewDesc.Format = textureDesc.Format;
renderTargetViewDesc.ViewDimension = D3D11_RTV_DIMENSION_TEXTURE2DARRAY;
renderTargetViewDesc.Texture2DArray.MipSlice = 0;
renderTargetViewDesc.Texture2DArray.FirstArraySlice =i;
renderTargetViewDesc.Texture2DArray.ArraySize = 1;

(注:如果您使用MSAA，ViewDimension将成为D3D11_RTV_DIMENSION_TEXTURE2DMSARRAY)

另外，要为资源创建着色器视图，因为您想要创建默认视图，所以可以传递NULL，而不是一个自制的D3D11_SHADER_RESOURCE_VIEW_DESC。

DX_CALL(DX11Internal::GetDevice()->CreateShaderResourceView(tex, NULL, &mSRV));

您还可以创建另一个呈现视图，该视图一次性映射整个多维数据集映射：

DX11Internal::GetDevice()->CreateRenderTargetView(tex, NULL, &cubeMapFullView);

但是在这种情况下，您需要使用几何图形着色器来跨片复制几何图形(使用SV_RenderTargetArrayIndex)，或者使用特定于供应商的扩展来提供相同的特性。