* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Document
System of linear equations wikipedia , lookup
Vector space wikipedia , lookup
Laplace–Runge–Lenz vector wikipedia , lookup
Jordan normal form wikipedia , lookup
Eigenvalues and eigenvectors wikipedia , lookup
Exterior algebra wikipedia , lookup
Rotation matrix wikipedia , lookup
Matrix (mathematics) wikipedia , lookup
Determinant wikipedia , lookup
Euclidean vector wikipedia , lookup
Gaussian elimination wikipedia , lookup
Perron–Frobenius theorem wikipedia , lookup
Non-negative matrix factorization wikipedia , lookup
Cross product wikipedia , lookup
Singular-value decomposition wikipedia , lookup
Covariance and contravariance of vectors wikipedia , lookup
Cayley–Hamilton theorem wikipedia , lookup
Orthogonal matrix wikipedia , lookup
Matrix calculus wikipedia , lookup
CSCE 590E Spring 2007 Basic Math By Jijun Tang Applied Trigonometry Trigonometric functions Defined using right triangle y sin a h h y a x x cos a h y sin a tan a x cos a Applied Trigonometry Angles measured in radians radians degrees p 180 180 p degrees radians Full circle contains 2p radians Trigonometry Trigonometric identities sin a sin a cos a sin a p 2 cos a cos a tan a tan a sin a cos a p 2 cos a sin a p 2 sin 2 a cos 2 a 1 sin a cos a p 2 sin a sin a p sin a p cos a cos a p cos a p Inverse trigonometric functions Return angle for which sin, cos, or tan function produces a particular value If sin a = z, then a = sin-1 z If cos a = z, then a = cos-1 z If tan a = z, then a = tan-1 z arcs Vectors and Matrices Scalars represent quantities that can be described fully using one value Mass Time Distance Vectors describe a magnitude and direction together using multiple values Vectors and Matrices Two vectors V and W are added by placing the beginning of W at the end of V Subtraction reverses the second vector W V V+W W V–W V V –W Vectors and Matrices Vectors add and subtract componentwise V W V1 W1 ,V2 W2 , ,Vn Wn V W V1 W1 ,V2 W2 , ,Vn Wn Vectors and Matrices The magnitude of an n-dimensional vector V is given by V n 2 V i i 1 In three dimensions, this is V Vx2 Vy2 Vz2 Vectors and Matrices A vector having a magnitude of 1 is called a unit vector Any vector V can be resized to unit length by dividing it by its magnitude: V ˆ V V This process is called normalization Vectors and Matrices A matrix is a rectangular array of numbers arranged as rows and columns A matrix having n rows and m columns is 1 2 3 an n m matrix M 4 5 6 At the right, M is a 2 3 matrix If n = m, the matrix is a square matrix Vectors and Matrices The transpose of a matrix M is denoted MT and has its rows and columns exchanged: 1 2 3 M 4 5 6 1 4 T M 2 5 3 6 Vectors and Matrices An n-dimensional vector V can be thought of as an n 1 column matrix: V V1 , V2 , V1 V2 , Vn Vn Or a 1 n row matrix: V T V1 V2 Vn Vectors and Matrices Product of two matrices A and B Number of columns of A must equal number of rows of B Entries of the product are given by m AB ij Aik Bkj k 1 If A is a n m matrix, and B is an m p matrix, then AB is an n p matrix Vectors and Matrices Example matrix product 2 3 2 1 8 13 M 1 1 4 5 6 6 M 11 2 2 3 4 M 12 2 1 3 5 13 8 M 21 1 2 1 4 6 M 22 1 1 1 5 6 Vectors and Matrices Matrices are used to transform vectors from one coordinate system to another In three dimensions, the product of a matrix and a column vector looks like: M 11 M 21 M 31 M 12 M 22 M 32 M 13 Vx M 11Vx M 12Vy M 13Vz M 23 Vy M 21Vx M 22Vy M 23Vz M 33 Vz M 31Vx M 32Vy M 33Vz Identity Matrix In For any n n matrix M, the product with the identity matrix is M itself I nM = M MIn = M Invertible An n n matrix M is invertible if there exists another matrix G such that 1 0 0 1 MG GM I n 0 0 0 0 1 The inverse of M is denoted M-1 Determinant The determinant of a square matrix M is denoted det M or |M| A matrix is invertible if its determinant is not zero For a 2 2 matrix, a b a b det ad bc c d c d Determinant The determinant of a 3 3 matrix is m11 m12 m13 m21 m22 m23 m11 m31 m32 m33 m22 m23 m32 m33 m12 m21 m23 m31 m33 m13 m21 m22 m31 m32 m11 m22 m33 m23 m32 m12 m21m33 m23m31 m13 m21m32 m22 m31 Inverse Explicit formulas exist for matrix inverses These are good for small matrices, but other methods are generally used for larger matrices In computer graphics, we are usually dealing with 2 2, 3 3, and a special form of 4 4 matrices Vectors and Matrices A special type of 4 4 matrix used in computer graphics looks like R11 R21 M R31 0 R12 R13 R22 R23 R32 R33 0 0 Tx Ty Tz 1 R is a 3 3 rotation matrix, and T is a translation vector Vectors and Matrices The inverse of this 4 4 matrix is M 1 R 1 0 R111 1 R T R211 R311 1 0 R121 R131 R221 R231 R321 R331 0 0 R 1T x 1 R T y R 1T z 1 The Dot Product The dot product is a product between two vectors that produces a scalar The dot product between two n-dimensional vectors V and W is n given by V W VW i i i 1 In three dimensions, V W VxWx VyWy VzWz The Dot Product The dot product can be used to project one vector onto another V a V cos a VW W W The Dot Product The dot product satisfies the formula V W V W cos a a is the angle between the two vectors Dot product is always 0 between perpendicular vectors If V and W are unit vectors, the dot product is 1 for parallel vectors pointing in the same direction, -1 for opposite The Dot Product The dot product of a vector with itself produces the squared magnitude VV V V V 2 Often, the notation V 2 is used as shorthand for V V The Cross Product The cross product is a product between two vectors the produces a vector The cross product only applies in three dimensions The cross product is perpendicular to both vectors being multiplied together The cross product between two parallel vectors is the zero vector (0, 0, 0) The Cross Product The cross product between V and W is V W VyWz VzWy , VzWx VxWz , VxWy VyWx A helpful tool for remembering this formula is the pseudodeterminant ˆi ˆj kˆ V W Vx Vy Vz Wx Wy Wz The Cross Product The cross product can also be expressed as the matrix-vector product 0 V W Vz Vy Vz 0 Vx Vy Wx Vx Wy 0 Wz The perpendicularity property means V W V 0 V W W 0 The Cross Product The cross product satisfies the trigonometric relationship V W V W sin a This is the area of the parallelogram formed by V V and W ||V|| sin a a W The Cross Product The area A of a triangle with vertices P1, P2, and P3 is thus given by A 1 P2 P1 P3 P1 2 The Cross Product Cross products obey the right hand rule If first vector points along right thumb, and second vector points along right fingers, Then cross product points out of right palm W V V W Reversing order of vectors negates the cross product: Transformations Calculations are often carried out in many different coordinate systems We must be able to transform information from one coordinate system to another easily Matrix multiplication allows us to do this Transformations Suppose that the coordinate axes in one coordinate system correspond to the directions R, S, and T in another Then we transform a vector V to the RST system as follows Rx W R S T V Ry Rz Sx Sy Sz Tx Vx Ty Vy Tz Vz ILLustration Transformation matrix We transform back to the original system by inverting the matrix: Rx V Ry Rz Sx Sy Sz 1 Tx Ty W Tz Often, the matrix’s inverse is equal to its transpose—such a matrix is called orthogonal Transformations A 3 3 matrix can reorient the coordinate axes in any way, but it leaves the origin fixed We must add a translation component D to move the origin: Rx W Ry Rz Sx Sy Sz Tx Vx Dx Ty Vy Dy Tz Vz Dz Transformations Homogeneous coordinates Four-dimensional space Combines 3 3 matrix and translation into one 4 4 matrix Rx Ry W Rz 0 Sx Tx Sy Ty Sz Tz 0 0 Dx Vx Dy Vy Dz Vz 1 Vw Transformations V is now a four-dimensional vector The w-coordinate of V determines whether V is a point or a direction vector If w = 0, then V is a direction vector and the fourth column of the transformation matrix has no effect If w 0, then V is a point and the fourth column of the matrix translates the origin Normally, w = 1 for points Transformations The three-dimensional counterpart of a four-dimensional homogeneous vector V is given by Vx Vy Vz V3D , , Vw Vw Vw Scaling a homogeneous vector thus has no effect on its actual 3D value Transformations Transformation matrices are often the result of combining several simple transformations Translations Scales Rotations Transformations are combined by multiplying their matrices together Transformation Steps Orderings Orderings Orderings of different type is important A rotation followed by a translation is different from a translation followed by a rotation Orderings of the same type does not matter Transformations Translation matrix M translate 1 0 0 0 0 0 Tx 1 0 Ty 0 1 Tz 0 0 1 Translates the origin by the vector T Transformations Scale matrix M scale a 0 0 0 0 0 0 b 0 0 0 c 0 0 0 1 Scales coordinate axes by a, b, and c If a = b = c, the scale is uniform Transformations Rotation matrix M z -rotate cos q sin q 0 0 sin q 0 0 cos q 0 0 0 1 0 0 0 1 Rotates points about the z-axis through the angle q Transformations Similar matrices for rotations about x, y M x -rotate M y -rotate 1 0 0 0 0 cos q sin q 0 0 sin q cos q 0 0 0 0 1 0 sin q 0 1 0 0 0 cos q 0 0 0 1 cos q 0 sin q 0 Transformations Normal vectors transform differently than do ordinary points and directions A normal vector represents the direction pointing out of a surface A normal vector is perpendicular to the tangent plane If a matrix M transforms points from one coordinate system to another, then normal vectors must be transformed by (M-1)T Geometry A line in 3D space is represented by P t S tV S is a point on the line, and V is the direction along which the line runs Any point P on the line corresponds to a value of the parameter t Two lines are parallel if their direction vectors are parallel Geometry A plane in 3D space can be defined by a normal direction N and a point P Other points in the plane satisfy N Q P 0 N Q P Geometry A plane equation is commonly written Ax By Cz D 0 A, B, and C are the components of the normal direction N, and D is given by D N P for any point P in the plane Geometry A plane is often represented by the 4D vector (A, B, C, D) If a 4D homogeneous point P lies in the plane, then (A, B, C, D) P = 0 If a point does not lie in the plane, then the dot product tells us which side of the plane the point lies on Geometry Distance d from a point P to a line S+tV P P S d S P S V V V Geometry Use Pythagorean theorem: P S V d P S V 2 2 Taking square root, d P S 2 2 P S V 2 V2 If V is unit length, then V 2 = 1 Geometry Intersection of a line and a plane Let P(t) = S + t V be the line Let L = (N, D) be the plane We want to find t such that L P(t) = 0 Lx S x Ly S y Lz S z Lw L S t LV LxVx LyVy LzVz Careful, S has w-coordinate of 1, and V has w-coordinate of 0 Geometry If L V = 0, the line is parallel to the plane and no intersection occurs Otherwise, the point of intersection is L S P t S V LV