/*========================================================================= Program: Visualization Toolkit Module: vtkPerspectiveTransform.h Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen All rights reserved. See Copyright.txt or http://www.kitware.com/Copyright.htm for details. This software is distributed WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the above copyright notice for more information. =========================================================================*/ /** * @class vtkPerspectiveTransform * @brief describes a 4x4 matrix transformation * * A vtkPerspectiveTransform can be used to describe the full range of * homogeneous transformations. It was designed in particular * to describe a camera-view of a scene. *

The order in which you set up the display coordinates (via * AdjustZBuffer() and AdjustViewport()), the projection (via Perspective(), * Frustum(), or Ortho()) and the camera view (via SetupCamera()) are * important. If the transform is in PreMultiply mode, which is the * default, set the Viewport and ZBuffer first, then the projection, and * finally the camera view. Once the view is set up, the Translate * and Rotate methods can be used to move the camera around in world * coordinates. If the Oblique() or Stereo() methods are used, they * should be called just before SetupCamera(). *

In PostMultiply mode, you must perform all transformations * in the opposite order. This is necessary, for example, if you * already have a perspective transformation set up but must adjust * the viewport. Another example is if you have a view transformation, * and wish to perform translations and rotations in the camera's * coordinate system rather than in world coordinates. *

The SetInput and Concatenate methods can be used to create * a transformation pipeline with vtkPerspectiveTransform. See vtkTransform * for more information on the transformation pipeline. * @sa * vtkGeneralTransform vtkTransform vtkMatrix4x4 vtkCamera */ #ifndef vtkPerspectiveTransform_h #define vtkPerspectiveTransform_h #include "vtkCommonTransformsModule.h" // For export macro #include "vtkHomogeneousTransform.h" #include "vtkMatrix4x4.h" // Needed for inline methods class VTKCOMMONTRANSFORMS_EXPORT vtkPerspectiveTransform : public vtkHomogeneousTransform { public: static vtkPerspectiveTransform* New(); vtkTypeMacro(vtkPerspectiveTransform, vtkHomogeneousTransform); void PrintSelf(ostream& os, vtkIndent indent) override; /** * Set this transformation to the identity transformation. If * the transform has an Input, then the transformation will be * reset so that it is the same as the Input. */ void Identity() { this->Concatenation->Identity(); this->Modified(); } /** * Invert the transformation. This will also set a flag so that * the transformation will use the inverse of its Input, if an Input * has been set. */ void Inverse() override { this->Concatenation->Inverse(); this->Modified(); } /** * Perform an adjustment to the viewport coordinates. By default Ortho, * Frustum, and Perspective provide a window of ([-1,+1],[-1,+1]). * In PreMultiply mode, you call this method before calling Ortho, Frustum, * or Perspective. In PostMultiply mode you can call it after. Note * that if you must apply both AdjustZBuffer and AdjustViewport, it * makes no difference which order you apply them in. */ void AdjustViewport(double oldXMin, double oldXMax, double oldYMin, double oldYMax, double newXMin, double newXMax, double newYMin, double newYMax); /** * Perform an adjustment to the Z-Buffer range that the near and far * clipping planes map to. By default Ortho, Frustum, and Perspective * map the near clipping plane to -1 and the far clipping plane to +1. * In PreMultiply mode, you call this method before calling Ortho, Frustum, * or Perspective. In PostMultiply mode you can call it after. */ void AdjustZBuffer(double oldNearZ, double oldFarZ, double newNearZ, double newFarZ); /** * Create an orthogonal projection matrix and concatenate it by the * current transformation. The matrix maps [xmin,xmax], [ymin,ymax], * [-znear,-zfar] to [-1,+1], [-1,+1], [+1,-1]. */ void Ortho(double xmin, double xmax, double ymin, double ymax, double znear, double zfar); /** * Create an perspective projection matrix and concatenate it by the * current transformation. The matrix maps a frustum with a back * plane at -zfar and a front plane at -znear with extent * [xmin,xmax],[ymin,ymax] to [-1,+1], [-1,+1], [+1,-1]. */ void Frustum(double xmin, double xmax, double ymin, double ymax, double znear, double zfar); /** * Create a perspective projection matrix by specifying the view angle * (this angle is in the y direction), the aspect ratio, and the near * and far clipping range. The projection matrix is concatenated * with the current transformation. This method works via Frustum. */ void Perspective(double angle, double aspect, double znear, double zfar); /** * Create a shear transformation about a plane at distance z from * the camera. The values dxdz (i.e. dx/dz) and dydz specify the * amount of shear in the x and y directions. The 'zplane' specifies * the distance from the camera to the plane at which the shear * causes zero displacement. Generally you want this plane to be the * focal plane. * This transformation can be used in combination with Ortho to create * an oblique projection. It can also be used in combination with * Perspective to provide correct stereo views when the eye is at * arbitrary but known positions relative to the center of a flat * viewing screen. */ void Shear(double dxdz, double dydz, double zplane); /** * Create a stereo shear matrix and concatenate it with the * current transformation. This can be applied in conjunction with either a * perspective transformation (via Frustum or Projection) or an * orthographic projection. You must specify the distance from * the camera plane to the focal plane, and the angle between * the distance vector and the eye. The angle should be negative * for the left eye, and positive for the right. This method * works via Oblique. */ void Stereo(double angle, double focaldistance); /** * Set a view transformation matrix for the camera (this matrix does * not contain any perspective) and concatenate it with the current * transformation. */ void SetupCamera(const double position[3], const double focalpoint[3], const double viewup[3]); void SetupCamera(double p0, double p1, double p2, double fp0, double fp1, double fp2, double vup0, double vup1, double vup2); //@{ /** * Create a translation matrix and concatenate it with the current * transformation according to PreMultiply or PostMultiply semantics. */ void Translate(double x, double y, double z) { this->Concatenation->Translate(x, y, z); } void Translate(const double x[3]) { this->Translate(x[0], x[1], x[2]); } void Translate(const float x[3]) { this->Translate(x[0], x[1], x[2]); } //@} //@{ /** * Create a rotation matrix and concatenate it with the current * transformation according to PreMultiply or PostMultiply semantics. * The angle is in degrees, and (x,y,z) specifies the axis that the * rotation will be performed around. */ void RotateWXYZ(double angle, double x, double y, double z) { this->Concatenation->Rotate(angle, x, y, z); } void RotateWXYZ(double angle, const double axis[3]) { this->RotateWXYZ(angle, axis[0], axis[1], axis[2]); } void RotateWXYZ(double angle, const float axis[3]) { this->RotateWXYZ(angle, axis[0], axis[1], axis[2]); } //@} //@{ /** * Create a rotation matrix about the X, Y, or Z axis and concatenate * it with the current transformation according to PreMultiply or * PostMultiply semantics. The angle is expressed in degrees. */ void RotateX(double angle) { this->RotateWXYZ(angle, 1, 0, 0); } void RotateY(double angle) { this->RotateWXYZ(angle, 0, 1, 0); } void RotateZ(double angle) { this->RotateWXYZ(angle, 0, 0, 1); } //@} //@{ /** * Create a scale matrix (i.e. set the diagonal elements to x, y, z) * and concatenate it with the current transformation according to * PreMultiply or PostMultiply semantics. */ void Scale(double x, double y, double z) { this->Concatenation->Scale(x, y, z); } void Scale(const double s[3]) { this->Scale(s[0], s[1], s[2]); } void Scale(const float s[3]) { this->Scale(s[0], s[1], s[2]); } //@} //@{ /** * Set the current matrix directly. This actually calls Identity(), * followed by Concatenate(matrix). */ void SetMatrix(vtkMatrix4x4* matrix) { this->SetMatrix(*matrix->Element); } void SetMatrix(const double elements[16]) { this->Identity(); this->Concatenate(elements); } //@} //@{ /** * Concatenates the matrix with the current transformation according * to PreMultiply or PostMultiply semantics. */ void Concatenate(vtkMatrix4x4* matrix) { this->Concatenate(*matrix->Element); } void Concatenate(const double elements[16]) { this->Concatenation->Concatenate(elements); } //@} /** * Concatenate the specified transform with the current transformation * according to PreMultiply or PostMultiply semantics. * The concatenation is pipelined, meaning that if any of the * transformations are changed, even after Concatenate() is called, * those changes will be reflected when you call TransformPoint(). */ void Concatenate(vtkHomogeneousTransform* transform); /** * Sets the internal state of the transform to PreMultiply. All subsequent * operations will occur before those already represented in the * current transformation. In homogeneous matrix notation, M = M*A where * M is the current transformation matrix and A is the applied matrix. * The default is PreMultiply. */ void PreMultiply() { if (this->Concatenation->GetPreMultiplyFlag()) { return; } this->Concatenation->SetPreMultiplyFlag(1); this->Modified(); } /** * Sets the internal state of the transform to PostMultiply. All subsequent * operations will occur after those already represented in the * current transformation. In homogeneous matrix notation, M = A*M where * M is the current transformation matrix and A is the applied matrix. * The default is PreMultiply. */ void PostMultiply() { if (!this->Concatenation->GetPreMultiplyFlag()) { return; } this->Concatenation->SetPreMultiplyFlag(0); this->Modified(); } /** * Get the total number of transformations that are linked into this * one via Concatenate() operations or via SetInput(). */ int GetNumberOfConcatenatedTransforms() { return this->Concatenation->GetNumberOfTransforms() + (this->Input == nullptr ? 0 : 1); } //@{ /** * Get one of the concatenated transformations as a vtkAbstractTransform. * These transformations are applied, in series, every time the * transformation of a coordinate occurs. This method is provided * to make it possible to decompose a transformation into its * constituents, for example to save a transformation to a file. */ vtkHomogeneousTransform* GetConcatenatedTransform(int i) { vtkAbstractTransform* t; if (this->Input == nullptr) { t = this->Concatenation->GetTransform(i); } else if (i < this->Concatenation->GetNumberOfPreTransforms()) { t = this->Concatenation->GetTransform(i); } else if (i > this->Concatenation->GetNumberOfPreTransforms()) { t = this->Concatenation->GetTransform(i - 1); } else if (this->GetInverseFlag()) { t = this->Input->GetInverse(); } else { t = this->Input; } return static_cast(t); } //@} //@{ /** * Set the input for this transformation. This will be used as the * base transformation if it is set. This method allows you to build * a transform pipeline: if the input is modified, then this transformation * will automatically update accordingly. Note that the InverseFlag, * controlled via Inverse(), determines whether this transformation * will use the Input or the inverse of the Input. */ void SetInput(vtkHomogeneousTransform* input); vtkHomogeneousTransform* GetInput() { return this->Input; } //@} /** * Get the inverse flag of the transformation. This controls * whether it is the Input or the inverse of the Input that * is used as the base transformation. The InverseFlag is * flipped every time Inverse() is called. The InverseFlag * is off when a transform is first created. */ int GetInverseFlag() { return this->Concatenation->GetInverseFlag(); } //@{ /** * Pushes the current transformation onto the transformation stack. */ void Push() { if (this->Stack == nullptr) { this->Stack = vtkTransformConcatenationStack::New(); } this->Stack->Push(&this->Concatenation); this->Modified(); } //@} //@{ /** * Deletes the transformation on the top of the stack and sets the top * to the next transformation on the stack. */ void Pop() { if (this->Stack == nullptr) { return; } this->Stack->Pop(&this->Concatenation); this->Modified(); } //@} /** * Make a new transform of the same type -- you are responsible for * deleting the transform when you are done with it. */ vtkAbstractTransform* MakeTransform() override; /** * Check for self-reference. Will return true if concatenating * with the specified transform, setting it to be our inverse, * or setting it to be our input will create a circular reference. * CircuitCheck is automatically called by SetInput(), SetInverse(), * and Concatenate(vtkXTransform *). Avoid using this function, * it is experimental. */ int CircuitCheck(vtkAbstractTransform* transform) override; /** * Override GetMTime to account for input and concatenation. */ vtkMTimeType GetMTime() override; protected: vtkPerspectiveTransform(); ~vtkPerspectiveTransform() override; void InternalDeepCopy(vtkAbstractTransform* t) override; void InternalUpdate() override; vtkHomogeneousTransform* Input; vtkTransformConcatenation* Concatenation; vtkTransformConcatenationStack* Stack; private: vtkPerspectiveTransform(const vtkPerspectiveTransform&) = delete; void operator=(const vtkPerspectiveTransform&) = delete; }; #endif