CameraResectioning.py
Go to the documentation of this file.
1 # pylint: disable=consider-using-from-import,invalid-name,no-name-in-module,no-member,missing-function-docstring
2 """
3 This is a 1:1 transcription of CameraResectioning.cpp.
4 """
5 import numpy as np
6 from gtsam import Cal3_S2, CustomFactor, LevenbergMarquardtOptimizer, KeyVector
7 from gtsam import NonlinearFactor, NonlinearFactorGraph
8 from gtsam import PinholeCameraCal3_S2, Point2, Point3, Pose3, Rot3, Values
9 from gtsam.noiseModel import Base as SharedNoiseModel, Diagonal
10 from gtsam.symbol_shorthand import X
11 
12 
14  model: SharedNoiseModel,
15  key: int,
16  calib: Cal3_S2,
17  p: Point2,
18  P: Point3,
19 ) -> NonlinearFactor:
20 
21  def error_func(this: CustomFactor, v: Values, H: list[np.ndarray]) -> np.ndarray:
22  pose = v.atPose3(this.keys()[0])
23  camera = PinholeCameraCal3_S2(pose, calib)
24  if H is None:
25  return camera.project(P) - p
26  Dpose = np.zeros((2, 6), order="F")
27  Dpoint = np.zeros((2, 3), order="F")
28  Dcal = np.zeros((2, 5), order="F")
29  result = camera.project(P, Dpose, Dpoint, Dcal) - p
30  H[0] = Dpose
31  return result
32 
33  return CustomFactor(model, KeyVector([key]), error_func)
34 
35 
36 def main() -> None:
37  """
38  Camera: f = 1, Image: 100x100, center: 50, 50.0
39  Pose (ground truth): (Xw, -Yw, -Zw, [0,0,2.0]')
40  Known landmarks:
41  3D Points: (10,10,0) (-10,10,0) (-10,-10,0) (10,-10,0)
42  Perfect measurements:
43  2D Point: (55,45) (45,45) (45,55) (55,55)
44  """
45 
46  # read camera intrinsic parameters
47  calib = Cal3_S2(1, 1, 0, 50, 50)
48 
49  # 1. create graph
50  graph = NonlinearFactorGraph()
51 
52  # 2. add factors to the graph
53  measurement_noise = Diagonal.Sigmas(np.array([0.5, 0.5]))
54  graph.add(
56  measurement_noise, X(1), calib, Point2(55, 45), Point3(10, 10, 0)
57  )
58  )
59  graph.add(
61  measurement_noise, X(1), calib, Point2(45, 45), Point3(-10, 10, 0)
62  )
63  )
64  graph.add(
66  measurement_noise, X(1), calib, Point2(45, 55), Point3(-10, -10, 0)
67  )
68  )
69  graph.add(
71  measurement_noise, X(1), calib, Point2(55, 55), Point3(10, -10, 0)
72  )
73  )
74 
75  # 3. Create an initial estimate for the camera pose
76  initial: Values = Values()
77  initial.insert(X(1), Pose3(Rot3(1, 0, 0, 0, -1, 0, 0, 0, -1), Point3(0, 0, 1)))
78 
79  # 4. Optimize the graph using Levenberg-Marquardt
80  result: Values = LevenbergMarquardtOptimizer(graph, initial).optimize()
81  result.print("Final result:\n")
82 
83 
84 if __name__ == "__main__":
85  main()
gtsam.examples.CameraResectioning.main
None main()
Definition: CameraResectioning.py:36
gtsam::optimize
Point3 optimize(const NonlinearFactorGraph &graph, const Values &values, Key landmarkKey)
Definition: triangulation.cpp:177
gtsam::CustomFactor
Definition: CustomFactor.h:45
gtsam::symbol_shorthand
Definition: inference/Symbol.h:147
X
#define X
Definition: icosphere.cpp:20
gtsam::KeyVector
FastVector< Key > KeyVector
Define collection type once and for all - also used in wrappers.
Definition: Key.h:92
gtsam.examples.CameraResectioning.resectioning_factor
NonlinearFactor resectioning_factor(SharedNoiseModel model, int key, Cal3_S2 calib, Point2 p, Point3 P)
Definition: CameraResectioning.py:13
gtsam::Rot3
Rot3 is a 3D rotation represented as a rotation matrix if the preprocessor symbol GTSAM_USE_QUATERNIO...
Definition: Rot3.h:58
gtsam::Pose3
Definition: Pose3.h:37
gtsam::PinholeCamera
Definition: PinholeCamera.h:33
gtsam::NonlinearFactorGraph
Definition: NonlinearFactorGraph.h:55
gtsam::LevenbergMarquardtOptimizer
Definition: LevenbergMarquardtOptimizer.h:35
gtsam::Point2
Vector2 Point2
Definition: Point2.h:32
gtsam::Values
Definition: Values.h:65
gtsam::noiseModel
All noise models live in the noiseModel namespace.
Definition: LossFunctions.cpp:28
gtsam::Cal3_S2
The most common 5DOF 3D->2D calibration.
Definition: Cal3_S2.h:34
gtsam::Point3
Vector3 Point3
Definition: Point3.h:38


gtsam
Author(s):
autogenerated on Sat Nov 16 2024 04:01:57