why normal transform use inverse-transpose matrix

定义$T$为切线,$N$为法线,$M$为切线的变换矩阵,$Q$为法线的变换矩阵

证明

$$
T\cdot N^T = 0\\[2ex]
T\cdot M = T’\\[2ex]
N\cdot Q = N’\\[2ex]
T’\cdot (N’)^T = 0\\[2ex]
T\cdot M \cdot (N\cdot Q)^T = 0\\[2ex]
T\cdot M \cdot Q^T \cdot N^T = 0\\[2ex]
\therefore M\cdot Q^T = I\\[2ex]
\therefore Q = (M^{-1})^T
$$

software recommend github page custom domain name

Comments

Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now

×