EuRoc数据集是有无人机采集的,它包括双目数据和IMU数据。本文以Machine Hall 01
为例,讲解它的双目数据如何使用。
1 左相机参数
cam0代表左相机,其相机内参为,
f
x
=
458.654
,
f
y
=
457.296
,
c
x
=
367.215
,
c
y
=
248.375
f_x = 458.654, f_y = 457.296, c_x = 367.215, c_y = 248.375
fx=458.654,fy=457.296,cx=367.215,cy=248.375
相机的畸变参数为,
k
1
=
−
0.28340811
,
k
2
=
0.07395907
,
p
1
=
0.00019359
,
p
2
=
1.76187114
×
1
0
−
5
k_1=-0.28340811, k_2=0.07395907, p_1=0.00019359, p_2=1.76187114 \times 10^{-5}
k1=−0.28340811,k2=0.07395907,p1=0.00019359,p2=1.76187114×10−5
相机畸变模型介绍如下,已知畸变前后相机归一化平面上的点分别记作
(
x
,
y
)
(x, y)
(x,y)和
(
x
d
i
s
t
o
r
t
e
d
,
y
d
i
s
t
o
r
t
e
d
)
(x_{distorted}, y_{distorted})
(xdistorted,ydistorted),且将
x
2
+
y
2
\sqrt{x^2+y^2}
x2+y2记作
r
r
r,那么有,
x
d
i
s
t
o
r
t
e
d
=
x
(
1
+
k
1
r
2
+
k
2
r
4
)
+
2
p
1
x
y
+
p
2
(
r
2
+
2
x
2
)
x_{distorted}=x(1+k_1r^2+k_2r^4)+2p_1xy+p_2(r^2+2x^2)
xdistorted=x(1+k1r2+k2r4)+2p1xy+p2(r2+2x2)
y d i s t o r t e d = y ( 1 + k 1 r 2 + k 2 r 4 ) + p 1 ( r 2 + 2 y 2 ) + 2 p 2 x y y_{distorted}=y(1+k_1r^2+k_2r^4)+p_1(r^2+2y^2)+2p_2xy ydistorted=y(1+k1r2+k2r4)+p1(r2+2y2)+2p2xy
2 右相机内参
cam1代表右相机,其相机内参为,
f
x
=
457.587
,
f
y
=
456.134
,
c
x
=
379.999
,
c
y
=
255.238
f_x = 457.587, f_y = 456.134, c_x = 379.999, c_y = 255.238
fx=457.587,fy=456.134,cx=379.999,cy=255.238
相机畸变参数为,
k
1
=
−
0.28368365
,
k
2
=
0.07451284
,
p
1
=
−
0.00010473
,
p
2
=
−
3.55590700
×
1
0
−
5
k_1 = -0.28368365, k_2 = 0.07451284, p_1 = -0.00010473, p_2 = -3.55590700 \times 10^{-5}
k1=−0.28368365,k2=0.07451284,p1=−0.00010473,p2=−3.55590700×10−5
3 左相机位姿真值
在state_groundtruth_estimate0
文件夹下存储了机体系的位姿真值,
更新中