Compare commits
58 Commits
6e151510cf
...
master
Author | SHA1 | Date | |
---|---|---|---|
43ac48bf56 | |||
e9d801bdc1 | |||
4a5bdc735c | |||
9253fcab5f | |||
69d385f257 | |||
62cd89fd0d | |||
a8090ca58a | |||
ed2ba61828 | |||
e94d4a06b5 | |||
0ef71b9220 | |||
5a80f97b68 | |||
876fa7617c | |||
14825c344e | |||
02156bd89a | |||
1380ec426f | |||
a7c296ca3d | |||
1a07ba3d3c | |||
3873c978dd | |||
6ee756941c | |||
6bcc72f826 | |||
737c23f32a | |||
58fe991647 | |||
d013a1b080 | |||
7b7e966217 | |||
53e2a5d524 | |||
3ae7bdb13a | |||
715ca6a6b4 | |||
9f528c1ea1 | |||
bfb2a551a7 | |||
af1820a238 | |||
655416a490 | |||
010db87e4b | |||
5451c2d097 | |||
b3e86b3e64 | |||
ebc73c0c5e | |||
273bbf8a94 | |||
1d6100ed13 | |||
6f7f8b7892 | |||
c565033d7f | |||
bcf948bcc1 | |||
02b9e0bc78 | |||
c79fc173b3 | |||
ecab936c62 | |||
2d045660c2 | |||
dcbc69a1c4 | |||
1b29047451 | |||
07f4927128 | |||
acca4ab018 | |||
44fffa19a2 | |||
9b4bf12846 | |||
5f6bcd1784 | |||
2a16fb2fc5 | |||
5d36a123a7 | |||
0d544c5361 | |||
d9899227c2 | |||
e788854fe8 | |||
4842c175a5 | |||
9ded72a366 |
16
.vscode/c_cpp_properties.json
vendored
Normal file
16
.vscode/c_cpp_properties.json
vendored
Normal file
@ -0,0 +1,16 @@
|
||||
{
|
||||
"configurations": [
|
||||
{
|
||||
"name": "Linux",
|
||||
"includePath": [
|
||||
"${workspaceFolder}/**"
|
||||
],
|
||||
"defines": [],
|
||||
"compilerPath": "/usr/bin/gcc",
|
||||
"cStandard": "c11",
|
||||
"cppStandard": "c++14",
|
||||
"intelliSenseMode": "clang-x64"
|
||||
}
|
||||
],
|
||||
"version": 4
|
||||
}
|
BIN
.vscode/ipch/778a17e566a4909e/mmap_address.bin
vendored
Normal file
BIN
.vscode/ipch/778a17e566a4909e/mmap_address.bin
vendored
Normal file
Binary file not shown.
BIN
.vscode/ipch/ccf983af1f87ec2b/mmap_address.bin
vendored
Normal file
BIN
.vscode/ipch/ccf983af1f87ec2b/mmap_address.bin
vendored
Normal file
Binary file not shown.
BIN
.vscode/ipch/e40aedd19a224f8d/mmap_address.bin
vendored
Normal file
BIN
.vscode/ipch/e40aedd19a224f8d/mmap_address.bin
vendored
Normal file
Binary file not shown.
6
.vscode/settings.json
vendored
Normal file
6
.vscode/settings.json
vendored
Normal file
@ -0,0 +1,6 @@
|
||||
{
|
||||
"files.associations": {
|
||||
"core": "cpp",
|
||||
"sparsecore": "cpp"
|
||||
}
|
||||
}
|
30
README.md
30
README.md
@ -1,12 +1,22 @@
|
||||
# MSCKF\_VIO
|
||||
|
||||
|
||||
The `MSCKF_VIO` package is a stereo version of MSCKF. The software takes in synchronized stereo images and IMU messages and generates real-time 6DOF pose estimation of the IMU frame.
|
||||
The `MSCKF_VIO` package is a stereo-photometric version of MSCKF. The software takes in synchronized stereo images and IMU messages and generates real-time 6DOF pose estimation of the IMU frame.
|
||||
|
||||
The software is tested on Ubuntu 16.04 with ROS Kinetic.
|
||||
This approach is based on the paper written by Ke Sun et al.
|
||||
[https://arxiv.org/abs/1712.00036](https://arxiv.org/abs/1712.00036) and their Stereo MSCKF implementation, which tightly fuse the matched feature information of a stereo image pair into a 6DOF Pose.
|
||||
The approach implemented in this repository follows the semi-dense msckf approach tightly fusing the photometric
|
||||
information around the matched featues into the covariance matrix, as described and derived in the master thesis [Pose Estimation using a Stereo-Photometric Multi-State Constraint Kalman Filter](http://raphael.maenle.net/resources/sp-msckf/maenle_master_thesis.pdf).
|
||||
|
||||
It's positioning is comparable to the approach from Ke Sun et al. with the photometric approach, with a higher
|
||||
computational load, especially with larger image patches around the feature. A video explaining the approach can be
|
||||
found on [https://youtu.be/HrqQywAnenQ](https://youtu.be/HrqQywAnenQ):
|
||||
<br/>
|
||||
[![Stereo-Photometric MSCKF](https://img.youtube.com/vi/HrqQywAnenQ/0.jpg)](https://www.youtube.com/watch?v=HrqQywAnenQ)
|
||||
|
||||
<br/>
|
||||
This software should be deployed using ROS Kinetic on Ubuntu 16.04 or 18.04.
|
||||
|
||||
Video: [https://www.youtube.com/watch?v=jxfJFgzmNSw&t](https://www.youtube.com/watch?v=jxfJFgzmNSw&t=3s)<br/>
|
||||
Paper Draft: [https://arxiv.org/abs/1712.00036](https://arxiv.org/abs/1712.00036)
|
||||
|
||||
## License
|
||||
|
||||
@ -28,16 +38,6 @@ cd your_work_space
|
||||
catkin_make --pkg msckf_vio --cmake-args -DCMAKE_BUILD_TYPE=Release
|
||||
```
|
||||
|
||||
## Calibration
|
||||
|
||||
An accurate calibration is crucial for successfully running the software. To get the best performance of the software, the stereo cameras and IMU should be hardware synchronized. Note that for the stereo calibration, which includes the camera intrinsics, distortion, and extrinsics between the two cameras, you have to use a calibration software. **Manually setting these parameters will not be accurate enough.** [Kalibr](https://github.com/ethz-asl/kalibr) can be used for the stereo calibration and also to get the transformation between the stereo cameras and IMU. The yaml file generated by Kalibr can be directly used in this software. See calibration files in the `config` folder for details. The two calibration files in the `config` folder should work directly with the EuRoC and [fast flight](https://github.com/KumarRobotics/msckf_vio/wiki) datasets. The convention of the calibration file is as follows:
|
||||
|
||||
`camx/T_cam_imu`: takes a vector from the IMU frame to the camx frame.
|
||||
`cam1/T_cn_cnm1`: takes a vector from the cam0 frame to the cam1 frame.
|
||||
|
||||
The filter uses the first 200 IMU messages to initialize the gyro bias, acc bias, and initial orientation. Therefore, the robot is required to start from a stationary state in order to initialize the VIO successfully.
|
||||
|
||||
|
||||
## EuRoC and UPenn Fast flight dataset example usage
|
||||
|
||||
First obtain either the [EuRoC](https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets) or the [UPenn fast flight](https://github.com/KumarRobotics/msckf_vio/wiki/Dataset) dataset.
|
||||
@ -75,6 +75,8 @@ To visualize the pose and feature estimates you can use the provided rviz config
|
||||
|
||||
## ROS Nodes
|
||||
|
||||
The general structure is similar to the structure of the MSCKF implementation this repository is derived from.
|
||||
|
||||
### `image_processor` node
|
||||
|
||||
**Subscribed Topics**
|
||||
|
@ -9,7 +9,7 @@ cam0:
|
||||
0, 0, 0, 1.000000000000000]
|
||||
camera_model: pinhole
|
||||
distortion_coeffs: [-0.28340811, 0.07395907, 0.00019359, 1.76187114e-05]
|
||||
distortion_model: radtan
|
||||
distortion_model: pre-radtan
|
||||
intrinsics: [458.654, 457.296, 367.215, 248.375]
|
||||
resolution: [752, 480]
|
||||
timeshift_cam_imu: 0.0
|
||||
@ -26,7 +26,7 @@ cam1:
|
||||
0, 0, 0, 1.000000000000000]
|
||||
camera_model: pinhole
|
||||
distortion_coeffs: [-0.28368365, 0.07451284, -0.00010473, -3.55590700e-05]
|
||||
distortion_model: radtan
|
||||
distortion_model: pre-radtan
|
||||
intrinsics: [457.587, 456.134, 379.999, 255.238]
|
||||
resolution: [752, 480]
|
||||
timeshift_cam_imu: 0.0
|
||||
|
36
config/camchain-imucam-tum-scaled.yaml
Normal file
36
config/camchain-imucam-tum-scaled.yaml
Normal file
@ -0,0 +1,36 @@
|
||||
cam0:
|
||||
T_cam_imu:
|
||||
[-0.9995378259923383, 0.02917807204183088, -0.008530798463872679, 0.047094247958417004,
|
||||
0.007526588843243184, -0.03435493139706542, -0.9993813532126198, -0.04788273017221637,
|
||||
-0.029453096117288798, -0.9989836729399656, 0.034119442089241274, -0.0697294754693238,
|
||||
0.0, 0.0, 0.0, 1.0]
|
||||
camera_model: pinhole
|
||||
distortion_coeffs: [0.0034823894022493434, 0.0007150348452162257, -0.0020532361418706202,
|
||||
0.00020293673591811182]
|
||||
distortion_model: pre-equidistant
|
||||
intrinsics: [190.97847715128717, 190.9733070521226, 254.93170605935475, 256.8974428996504]
|
||||
resolution: [512, 512]
|
||||
rostopic: /cam0/image_raw
|
||||
cam1:
|
||||
T_cam_imu:
|
||||
[-0.9995240747493029, 0.02986739485347808, -0.007717688852024281, -0.05374086123613335,
|
||||
0.008095979457928231, 0.01256553460985914, -0.9998882749870535, -0.04648588412432889,
|
||||
-0.02976708103202316, -0.9994748851595197, -0.0128013601698453, -0.07333210787623645,
|
||||
0.0, 0.0, 0.0, 1.0]
|
||||
T_cn_cnm1:
|
||||
[0.9999994317488622, -0.0008361847221513937, -0.0006612844045898121, -0.10092123225528335,
|
||||
0.0008042457277382264, 0.9988989443471681, -0.04690684567228134, -0.001964540595211977,
|
||||
0.0006997790813734836, 0.04690628718225568, 0.9988990492196964, -0.0014663556043866572,
|
||||
0.0, 0.0, 0.0, 1.0]
|
||||
camera_model: pinhole
|
||||
distortion_coeffs: [0.0034003170790442797, 0.001766278153469831, -0.00266312569781606,
|
||||
0.0003299517423931039]
|
||||
distortion_model: pre-equidistant
|
||||
intrinsics: [190.44236969414825, 190.4344384721956, 252.59949716835982, 254.91723064636983]
|
||||
resolution: [512, 512]
|
||||
rostopic: /cam1/image_raw
|
||||
T_imu_body:
|
||||
[1.0000, 0.0000, 0.0000, 0.0000,
|
||||
0.0000, 1.0000, 0.0000, 0.0000,
|
||||
0.0000, 0.0000, 1.0000, 0.0000,
|
||||
0.0000, 0.0000, 0.0000, 1.0000]
|
@ -18,6 +18,8 @@ namespace msckf_vio {
|
||||
|
||||
struct Frame{
|
||||
cv::Mat image;
|
||||
cv::Mat dximage;
|
||||
cv::Mat dyimage;
|
||||
double exposureTime_ms;
|
||||
};
|
||||
|
||||
@ -39,6 +41,7 @@ struct CameraCalibration{
|
||||
cv::Vec4d distortion_coeffs;
|
||||
movingWindow moving_window;
|
||||
cv::Mat featureVisu;
|
||||
int id;
|
||||
};
|
||||
|
||||
|
||||
|
@ -70,6 +70,11 @@ struct Feature {
|
||||
position(Eigen::Vector3d::Zero()),
|
||||
is_initialized(false), is_anchored(false) {}
|
||||
|
||||
|
||||
void Rhocost(const Eigen::Isometry3d& T_c0_ci,
|
||||
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
|
||||
double& e) const;
|
||||
|
||||
/*
|
||||
* @brief cost Compute the cost of the camera observations
|
||||
* @param T_c0_c1 A rigid body transformation takes
|
||||
@ -82,6 +87,13 @@ struct Feature {
|
||||
const Eigen::Vector3d& x, const Eigen::Vector2d& z,
|
||||
double& e) const;
|
||||
|
||||
bool initializeRho(const CamStateServer& cam_states);
|
||||
|
||||
inline void RhoJacobian(const Eigen::Isometry3d& T_c0_ci,
|
||||
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
|
||||
Eigen::Matrix<double, 2, 1>& J, Eigen::Vector2d& r,
|
||||
double& w) const;
|
||||
|
||||
/*
|
||||
* @brief jacobian Compute the Jacobian of the camera observation
|
||||
* @param T_c0_c1 A rigid body transformation takes
|
||||
@ -97,6 +109,10 @@ struct Feature {
|
||||
Eigen::Matrix<double, 2, 3>& J, Eigen::Vector2d& r,
|
||||
double& w) const;
|
||||
|
||||
inline double generateInitialDepth(
|
||||
const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1,
|
||||
const Eigen::Vector2d& z2) const;
|
||||
|
||||
/*
|
||||
* @brief generateInitialGuess Compute the initial guess of
|
||||
* the feature's 3d position using only two views.
|
||||
@ -122,14 +138,6 @@ struct Feature {
|
||||
const CamStateServer& cam_states) const;
|
||||
|
||||
|
||||
/*
|
||||
* @brief AnchorPixelToPosition projects an undistorted point in the
|
||||
* anchor frame back into the real world using the rho calculated
|
||||
* based on featur position
|
||||
*/
|
||||
inline Eigen::Vector3d AnchorPixelToPosition(cv::Point2f in_p, const CameraCalibration& cam);
|
||||
|
||||
|
||||
/*
|
||||
* @brief InitializeAnchor generates the NxN patch around the
|
||||
* feature in the Anchor image
|
||||
@ -157,8 +165,7 @@ struct Feature {
|
||||
inline bool initializePosition(
|
||||
const CamStateServer& cam_states);
|
||||
|
||||
|
||||
cv::Point2f pixelDistanceAt(
|
||||
cv::Point2f pixelDistanceAt(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
const CameraCalibration& cam,
|
||||
@ -177,6 +184,21 @@ cv::Point2f pixelDistanceAt(
|
||||
const CameraCalibration& cam,
|
||||
Eigen::Vector3d& in_p) const;
|
||||
|
||||
double CompleteCvKernel(
|
||||
const cv::Point2f pose,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam,
|
||||
std::string type) const;
|
||||
|
||||
double cvKernel(
|
||||
const cv::Point2f pose,
|
||||
std::string type) const;
|
||||
|
||||
double Kernel(
|
||||
const cv::Point2f pose,
|
||||
const cv::Mat frame,
|
||||
std::string type) const;
|
||||
|
||||
/*
|
||||
* @brief IrradianceAnchorPatch_Camera returns irradiance values
|
||||
* of the Anchor Patch position in a camera frame
|
||||
@ -186,32 +208,31 @@ cv::Point2f pixelDistanceAt(
|
||||
bool estimate_FrameIrradiance(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam0,
|
||||
CameraCalibration& cam,
|
||||
std::vector<double>& anchorPatch_estimate,
|
||||
IlluminationParameter& estimatedIllumination) const;
|
||||
|
||||
bool MarkerGeneration(
|
||||
bool MarkerGeneration(
|
||||
ros::Publisher& marker_pub,
|
||||
const CamStateServer& cam_states) const;
|
||||
|
||||
bool VisualizeKernel(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam0) const;
|
||||
|
||||
bool VisualizePatch(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam0,
|
||||
const std::vector<double> photo_r,
|
||||
CameraCalibration& cam,
|
||||
const Eigen::VectorXd& photo_r,
|
||||
std::stringstream& ss) const;
|
||||
|
||||
|
||||
/* @brief takes a pure pixel position (1m from image)
|
||||
* converts to actual pixel value and returns patch irradiance
|
||||
* around this pixel
|
||||
/*
|
||||
* @brief AnchorPixelToPosition uses the calcualted pixels
|
||||
* of the anchor patch to generate 3D positions of all of em
|
||||
*/
|
||||
void PatchAroundPurePixel(cv::Point2f p,
|
||||
int N,
|
||||
const CameraCalibration& cam,
|
||||
const StateIDType& cam_state_id,
|
||||
std::vector<float>& return_i) const;
|
||||
|
||||
inline Eigen::Vector3d AnchorPixelToPosition(cv::Point2f in_p,
|
||||
const CameraCalibration& cam);
|
||||
|
||||
/*
|
||||
* @brief Irradiance returns irradiance value of a pixel
|
||||
@ -256,6 +277,14 @@ cv::Point2f pixelDistanceAt(
|
||||
bool is_initialized;
|
||||
bool is_anchored;
|
||||
|
||||
|
||||
cv::Mat abs_xderImage;
|
||||
cv::Mat abs_yderImage;
|
||||
|
||||
cv::Mat xderImage;
|
||||
cv::Mat yderImage;
|
||||
|
||||
cv::Mat anchorImage_blurred;
|
||||
cv::Point2f anchor_center_pos;
|
||||
cv::Point2f undist_anchor_center_pos;
|
||||
// Noise for a normalized feature measurement.
|
||||
@ -271,6 +300,26 @@ typedef std::map<FeatureIDType, Feature, std::less<int>,
|
||||
Eigen::aligned_allocator<
|
||||
std::pair<const FeatureIDType, Feature> > > MapServer;
|
||||
|
||||
void Feature::Rhocost(const Eigen::Isometry3d& T_c0_ci,
|
||||
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
|
||||
double& e) const
|
||||
{
|
||||
// Compute hi1, hi2, and hi3 as Equation (37).
|
||||
const double& rho = x;
|
||||
|
||||
Eigen::Vector3d h = T_c0_ci.linear()*
|
||||
Eigen::Vector3d(z1(0), z1(1), 1.0) + rho*T_c0_ci.translation();
|
||||
double& h1 = h(0);
|
||||
double& h2 = h(1);
|
||||
double& h3 = h(2);
|
||||
|
||||
// Predict the feature observation in ci frame.
|
||||
Eigen::Vector2d z_hat(h1/h3, h2/h3);
|
||||
|
||||
// Compute the residual.
|
||||
e = (z_hat-z2).squaredNorm();
|
||||
return;
|
||||
}
|
||||
|
||||
void Feature::cost(const Eigen::Isometry3d& T_c0_ci,
|
||||
const Eigen::Vector3d& x, const Eigen::Vector2d& z,
|
||||
@ -283,9 +332,9 @@ void Feature::cost(const Eigen::Isometry3d& T_c0_ci,
|
||||
|
||||
Eigen::Vector3d h = T_c0_ci.linear()*
|
||||
Eigen::Vector3d(alpha, beta, 1.0) + rho*T_c0_ci.translation();
|
||||
double& h1 = h(0);
|
||||
double& h2 = h(1);
|
||||
double& h3 = h(2);
|
||||
double h1 = h(0);
|
||||
double h2 = h(1);
|
||||
double h3 = h(2);
|
||||
|
||||
// Predict the feature observation in ci frame.
|
||||
Eigen::Vector2d z_hat(h1/h3, h2/h3);
|
||||
@ -295,6 +344,42 @@ void Feature::cost(const Eigen::Isometry3d& T_c0_ci,
|
||||
return;
|
||||
}
|
||||
|
||||
void Feature::RhoJacobian(const Eigen::Isometry3d& T_c0_ci,
|
||||
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
|
||||
Eigen::Matrix<double, 2, 1>& J, Eigen::Vector2d& r,
|
||||
double& w) const
|
||||
{
|
||||
|
||||
const double& rho = x;
|
||||
|
||||
Eigen::Vector3d h = T_c0_ci.linear()*
|
||||
Eigen::Vector3d(z1(0), z2(1), 1.0) + rho*T_c0_ci.translation();
|
||||
double& h1 = h(0);
|
||||
double& h2 = h(1);
|
||||
double& h3 = h(2);
|
||||
|
||||
// Compute the Jacobian.
|
||||
Eigen::Matrix3d W;
|
||||
W.leftCols<2>() = T_c0_ci.linear().leftCols<2>();
|
||||
W.rightCols<1>() = T_c0_ci.translation();
|
||||
|
||||
J(0,0) = -h1/(h3*h3);
|
||||
J(1,0) = -h2/(h3*h3);
|
||||
|
||||
// Compute the residual.
|
||||
Eigen::Vector2d z_hat(h1/h3, h2/h3);
|
||||
r = z_hat - z2;
|
||||
|
||||
// Compute the weight based on the residual.
|
||||
double e = r.norm();
|
||||
if (e <= optimization_config.huber_epsilon)
|
||||
w = 1.0;
|
||||
else
|
||||
w = optimization_config.huber_epsilon / (2*e);
|
||||
|
||||
return;
|
||||
}
|
||||
|
||||
void Feature::jacobian(const Eigen::Isometry3d& T_c0_ci,
|
||||
const Eigen::Vector3d& x, const Eigen::Vector2d& z,
|
||||
Eigen::Matrix<double, 2, 3>& J, Eigen::Vector2d& r,
|
||||
@ -334,9 +419,9 @@ void Feature::jacobian(const Eigen::Isometry3d& T_c0_ci,
|
||||
return;
|
||||
}
|
||||
|
||||
void Feature::generateInitialGuess(
|
||||
double Feature::generateInitialDepth(
|
||||
const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1,
|
||||
const Eigen::Vector2d& z2, Eigen::Vector3d& p) const
|
||||
const Eigen::Vector2d& z2) const
|
||||
{
|
||||
// Construct a least square problem to solve the depth.
|
||||
Eigen::Vector3d m = T_c1_c2.linear() * Eigen::Vector3d(z1(0), z1(1), 1.0);
|
||||
@ -351,6 +436,15 @@ void Feature::generateInitialGuess(
|
||||
|
||||
// Solve for the depth.
|
||||
double depth = (A.transpose() * A).inverse() * A.transpose() * b;
|
||||
return depth;
|
||||
}
|
||||
|
||||
|
||||
void Feature::generateInitialGuess(
|
||||
const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1,
|
||||
const Eigen::Vector2d& z2, Eigen::Vector3d& p) const
|
||||
{
|
||||
double depth = generateInitialDepth(T_c1_c2, z1, z2);
|
||||
p(0) = z1(0) * depth;
|
||||
p(1) = z1(1) * depth;
|
||||
p(2) = depth;
|
||||
@ -400,10 +494,62 @@ bool Feature::checkMotion(const CamStateServer& cam_states) const
|
||||
else return false;
|
||||
}
|
||||
|
||||
double Feature::CompleteCvKernel(
|
||||
const cv::Point2f pose,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam,
|
||||
std::string type) const
|
||||
{
|
||||
|
||||
double delta = 0;
|
||||
|
||||
if(type == "Sobel_x")
|
||||
delta = ((double)cam.moving_window.find(cam_state_id)->second.dximage.at<short>(pose.y, pose.x))/255.;
|
||||
else if (type == "Sobel_y")
|
||||
delta = ((double)cam.moving_window.find(cam_state_id)->second.dyimage.at<short>(pose.y, pose.x))/255.;
|
||||
|
||||
return delta;
|
||||
}
|
||||
|
||||
double Feature::cvKernel(
|
||||
const cv::Point2f pose,
|
||||
std::string type) const
|
||||
{
|
||||
|
||||
double delta = 0;
|
||||
|
||||
if(type == "Sobel_x")
|
||||
delta = ((double)xderImage.at<short>(pose.y, pose.x))/255.;
|
||||
else if (type == "Sobel_y")
|
||||
delta = ((double)yderImage.at<short>(pose.y, pose.x))/255.;
|
||||
return delta;
|
||||
|
||||
}
|
||||
|
||||
double Feature::Kernel(
|
||||
const cv::Point2f pose,
|
||||
const cv::Mat frame,
|
||||
std::string type) const
|
||||
{
|
||||
Eigen::Matrix<double, 3, 3> kernel = Eigen::Matrix<double, 3, 3>::Zero();
|
||||
if(type == "Sobel_x")
|
||||
kernel << -1., 0., 1.,-2., 0., 2. , -1., 0., 1.;
|
||||
else if(type == "Sobel_y")
|
||||
kernel << -1., -2., -1., 0., 0., 0., 1., 2., 1.;
|
||||
|
||||
double delta = 0;
|
||||
int offs = (int)(kernel.rows()-1)/2;
|
||||
|
||||
for(int i = 0; i < kernel.rows(); i++)
|
||||
for(int j = 0; j < kernel.cols(); j++)
|
||||
delta += ((float)frame.at<uint8_t>(pose.y+j-offs , pose.x+i-offs))/255. * (float)kernel(j,i);
|
||||
|
||||
return delta;
|
||||
}
|
||||
bool Feature::estimate_FrameIrradiance(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam0,
|
||||
CameraCalibration& cam,
|
||||
std::vector<double>& anchorPatch_estimate,
|
||||
IlluminationParameter& estimated_illumination) const
|
||||
{
|
||||
@ -412,14 +558,11 @@ bool Feature::estimate_FrameIrradiance(
|
||||
// muliply by a and add b of this frame
|
||||
|
||||
auto anchor = observations.begin();
|
||||
if(cam0.moving_window.find(anchor->first) == cam0.moving_window.end())
|
||||
{
|
||||
std::cout << "anchor not in buffer anymore!" << std::endl;
|
||||
if(cam.moving_window.find(anchor->first) == cam.moving_window.end())
|
||||
return false;
|
||||
}
|
||||
|
||||
double anchorExposureTime_ms = cam0.moving_window.find(anchor->first)->second.exposureTime_ms;
|
||||
double frameExposureTime_ms = cam0.moving_window.find(cam_state_id)->second.exposureTime_ms;
|
||||
double anchorExposureTime_ms = cam.moving_window.find(anchor->first)->second.exposureTime_ms;
|
||||
double frameExposureTime_ms = cam.moving_window.find(cam_state_id)->second.exposureTime_ms;
|
||||
|
||||
|
||||
double a_A = anchorExposureTime_ms;
|
||||
@ -554,11 +697,47 @@ bool Feature::MarkerGeneration(
|
||||
marker_pub.publish(ma);
|
||||
}
|
||||
|
||||
|
||||
bool Feature::VisualizeKernel(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam0) const
|
||||
{
|
||||
auto anchor = observations.begin();
|
||||
cv::Mat anchorImage = cam0.moving_window.find(anchor->first)->second.image;
|
||||
|
||||
//cv::Mat xderImage;
|
||||
//cv::Mat yderImage;
|
||||
|
||||
//cv::Sobel(anchorImage, xderImage, CV_8UC1, 1, 0, 3);
|
||||
//cv::Sobel(anchorImage, yderImage, CV_8UC1, 0, 1, 3);
|
||||
|
||||
|
||||
// cv::Mat xderImage2(anchorImage.rows, anchorImage.cols, anchorImage_blurred.type());
|
||||
// cv::Mat yderImage2(anchorImage.rows, anchorImage.cols, anchorImage_blurred.type());
|
||||
|
||||
/*
|
||||
for(int i = 1; i < anchorImage.rows-1; i++)
|
||||
for(int j = 1; j < anchorImage.cols-1; j++)
|
||||
xderImage2.at<uint8_t>(j,i) = 255.*fabs(Kernel(cv::Point2f(i,j), anchorImage_blurred, "Sobel_x"));
|
||||
|
||||
|
||||
for(int i = 1; i < anchorImage.rows-1; i++)
|
||||
for(int j = 1; j < anchorImage.cols-1; j++)
|
||||
yderImage2.at<uint8_t>(j,i) = 255.*fabs(Kernel(cv::Point2f(i,j), anchorImage_blurred, "Sobel_y"));
|
||||
*/
|
||||
//cv::imshow("anchor", anchorImage);
|
||||
cv::imshow("xder2", xderImage);
|
||||
cv::imshow("yder2", yderImage);
|
||||
|
||||
cvWaitKey(0);
|
||||
}
|
||||
|
||||
bool Feature::VisualizePatch(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
CameraCalibration& cam0,
|
||||
const std::vector<double> photo_r,
|
||||
CameraCalibration& cam,
|
||||
const Eigen::VectorXd& photo_r,
|
||||
std::stringstream& ss) const
|
||||
{
|
||||
|
||||
@ -566,7 +745,7 @@ bool Feature::VisualizePatch(
|
||||
|
||||
//visu - anchor
|
||||
auto anchor = observations.begin();
|
||||
cv::Mat anchorImage = cam0.moving_window.find(anchor->first)->second.image;
|
||||
cv::Mat anchorImage = cam.moving_window.find(anchor->first)->second.image;
|
||||
cv::Mat dottedFrame(anchorImage.size(), CV_8UC3);
|
||||
cv::cvtColor(anchorImage, dottedFrame, CV_GRAY2RGB);
|
||||
|
||||
@ -578,10 +757,10 @@ bool Feature::VisualizePatch(
|
||||
cv::Point ys(point.x, point.y);
|
||||
cv::rectangle(dottedFrame, xs, ys, cv::Scalar(0,255,255));
|
||||
}
|
||||
cam0.featureVisu = dottedFrame.clone();
|
||||
cam.featureVisu = dottedFrame.clone();
|
||||
|
||||
// visu - feature
|
||||
cv::Mat current_image = cam0.moving_window.find(cam_state_id)->second.image;
|
||||
cv::Mat current_image = cam.moving_window.find(cam_state_id)->second.image;
|
||||
cv::cvtColor(current_image, dottedFrame, CV_GRAY2RGB);
|
||||
|
||||
// set position in frame
|
||||
@ -589,7 +768,7 @@ bool Feature::VisualizePatch(
|
||||
std::vector<double> projectionPatch;
|
||||
for(auto point : anchorPatch_3d)
|
||||
{
|
||||
cv::Point2f p_in_c0 = projectPositionToCamera(cam_state, cam_state_id, cam0, point);
|
||||
cv::Point2f p_in_c0 = projectPositionToCamera(cam_state, cam_state_id, cam, point);
|
||||
projectionPatch.push_back(PixelIrradiance(p_in_c0, current_image));
|
||||
// visu - feature
|
||||
cv::Point xs(p_in_c0.x, p_in_c0.y);
|
||||
@ -597,7 +776,7 @@ bool Feature::VisualizePatch(
|
||||
cv::rectangle(dottedFrame, xs, ys, cv::Scalar(0,255,0));
|
||||
}
|
||||
|
||||
cv::hconcat(cam0.featureVisu, dottedFrame, cam0.featureVisu);
|
||||
cv::hconcat(cam.featureVisu, dottedFrame, cam.featureVisu);
|
||||
|
||||
|
||||
// patches visualization
|
||||
@ -622,9 +801,10 @@ bool Feature::VisualizePatch(
|
||||
CV_FILLED);
|
||||
|
||||
// irradiance grid projection
|
||||
|
||||
namer.str(std::string());
|
||||
namer << "projection";
|
||||
cv::putText(irradianceFrame, namer.str(), cvPoint(30, 45+scale*N),
|
||||
cv::putText(irradianceFrame, namer.str() , cvPoint(30, 45+scale*N),
|
||||
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
|
||||
|
||||
for(int i = 0; i<N; i++)
|
||||
@ -643,10 +823,14 @@ bool Feature::VisualizePatch(
|
||||
cv::putText(irradianceFrame, namer.str() , cvPoint(30, 65+scale*2*N),
|
||||
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
|
||||
|
||||
cv::Point2f p_f(observations.find(cam_state_id)->second(0),observations.find(cam_state_id)->second(1));
|
||||
cv::Point2f p_f;
|
||||
if(cam.id == 0)
|
||||
p_f = cv::Point2f(observations.find(cam_state_id)->second(0),observations.find(cam_state_id)->second(1));
|
||||
else if(cam.id == 1)
|
||||
p_f = cv::Point2f(observations.find(cam_state_id)->second(2),observations.find(cam_state_id)->second(3));
|
||||
// move to real pixels
|
||||
p_f = image_handler::distortPoint(p_f, cam0.intrinsics, cam0.distortion_model, cam0.distortion_coeffs);
|
||||
|
||||
p_f = image_handler::distortPoint(p_f, cam.intrinsics, cam.distortion_model, cam.distortion_coeffs);
|
||||
for(int i = 0; i<N; i++)
|
||||
{
|
||||
for(int j = 0; j<N ; j++)
|
||||
@ -663,22 +847,23 @@ bool Feature::VisualizePatch(
|
||||
// residual grid projection, positive - red, negative - blue colored
|
||||
namer.str(std::string());
|
||||
namer << "residual";
|
||||
std::cout << "-- photo_r -- \n" << photo_r << " -- " << std::endl;
|
||||
cv::putText(irradianceFrame, namer.str() , cvPoint(30+scale*N, scale*N/2-5),
|
||||
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
|
||||
|
||||
for(int i = 0; i<N; i++)
|
||||
for(int j = 0; j<N; j++)
|
||||
if(photo_r[i*N+j]>0)
|
||||
if(photo_r(2*(i*N+j))>0)
|
||||
cv::rectangle(irradianceFrame,
|
||||
cv::Point(40+scale*(N+i+1), 15+scale*(N/2+j)),
|
||||
cv::Point(40+scale*(N+i), 15+scale*(N/2+j+1)),
|
||||
cv::Scalar(255 - photo_r[i*N+j]*255, 255 - photo_r[i*N+j]*255, 255),
|
||||
cv::Scalar(255 - photo_r(2*(i*N+j)+1)*255, 255 - photo_r(2*(i*N+j)+1)*255, 255),
|
||||
CV_FILLED);
|
||||
else
|
||||
cv::rectangle(irradianceFrame,
|
||||
cv::Point(40+scale*(N+i+1), 15+scale*(N/2+j)),
|
||||
cv::Point(40+scale*(N+i), 15+scale*(N/2+j+1)),
|
||||
cv::Scalar(255, 255 + photo_r[i*N+j]*255, 255 + photo_r[i*N+j]*255),
|
||||
cv::Scalar(255, 255 + photo_r(2*(i*N+j)+1)*255, 255 + photo_r(2*(i*N+j)+1)*255),
|
||||
CV_FILLED);
|
||||
|
||||
// gradient arrow
|
||||
@ -688,22 +873,21 @@ bool Feature::VisualizePatch(
|
||||
cv::Point(30+scale*(N/2+0.5)+scale*gradientVector.x, 50+scale*(N+(N/2)+0.5)+scale*gradientVector.y),
|
||||
cv::Scalar(100, 0, 255),
|
||||
1);
|
||||
*/
|
||||
|
||||
// residual gradient direction
|
||||
/*
|
||||
cv::arrowedLine(irradianceFrame,
|
||||
cv::Point(40+scale*(N+N/2+0.5), 15+scale*((N-0.5))),
|
||||
cv::Point(40+scale*(N+N/2+0.5)+scale*residualVector.x, 15+scale*(N-0.5)+scale*residualVector.y),
|
||||
cv::Scalar(0, 255, 175),
|
||||
3);
|
||||
*/
|
||||
|
||||
*/
|
||||
cv::hconcat(cam0.featureVisu, irradianceFrame, cam0.featureVisu);
|
||||
|
||||
|
||||
cv::hconcat(cam.featureVisu, irradianceFrame, cam.featureVisu);
|
||||
|
||||
/*
|
||||
// visualize position of used observations and resulting feature position
|
||||
/*
|
||||
|
||||
cv::Mat positionFrame(anchorImage.size(), CV_8UC3, cv::Scalar(255, 240, 255));
|
||||
cv::resize(positionFrame, positionFrame, cv::Size(), rescale, rescale);
|
||||
|
||||
@ -732,17 +916,15 @@ bool Feature::VisualizePatch(
|
||||
|
||||
// draw, x y position and arrow with direction - write z next to it
|
||||
|
||||
cv::resize(cam0.featureVisu, cam0.featureVisu, cv::Size(), rescale, rescale);
|
||||
cv::resize(cam.featureVisu, cam.featureVisu, cv::Size(), rescale, rescale);
|
||||
|
||||
|
||||
cv::hconcat(cam0.featureVisu, positionFrame, cam0.featureVisu);
|
||||
|
||||
*/
|
||||
|
||||
cv::hconcat(cam.featureVisu, positionFrame, cam.featureVisu);
|
||||
*/
|
||||
// write feature position
|
||||
std::stringstream pos_s;
|
||||
pos_s << "u: " << observations.begin()->second(0) << " v: " << observations.begin()->second(1);
|
||||
cv::putText(cam0.featureVisu, ss.str() , cvPoint(anchorImage.rows + 100, anchorImage.cols - 30),
|
||||
cv::putText(cam.featureVisu, ss.str() , cvPoint(anchorImage.rows + 100, anchorImage.cols - 30),
|
||||
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(200,200,250), 1, CV_AA);
|
||||
// create line?
|
||||
|
||||
@ -750,35 +932,16 @@ bool Feature::VisualizePatch(
|
||||
|
||||
std::stringstream loc;
|
||||
// loc << "/home/raphael/dev/MSCKF_ws/img/feature_" << std::to_string(ros::Time::now().toSec()) << ".jpg";
|
||||
//cv::imwrite(loc.str(), cam0.featureVisu);
|
||||
|
||||
cv::imshow("patch", cam0.featureVisu);
|
||||
cvWaitKey(0);
|
||||
}
|
||||
|
||||
void Feature::PatchAroundPurePixel(cv::Point2f p,
|
||||
int N,
|
||||
const CameraCalibration& cam,
|
||||
const StateIDType& cam_state_id,
|
||||
std::vector<float>& return_i) const
|
||||
{
|
||||
int n = (int)(N-1)/2;
|
||||
cv::Mat image = cam.moving_window.find(cam_state_id)->second.image;
|
||||
cv::Point2f img_p = image_handler::distortPoint(p,
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs);
|
||||
|
||||
for(double u_run = -n; u_run <= n; u_run++)
|
||||
for(double v_run = -n; v_run <= n; v_run++)
|
||||
return_i.push_back(PixelIrradiance(cv::Point2f(img_p.x+u_run, img_p.y+v_run), image));
|
||||
//cv::imwrite(loc.str(), cam.featureVisu);
|
||||
|
||||
cv::imshow("patch", cam.featureVisu);
|
||||
cvWaitKey(1);
|
||||
}
|
||||
|
||||
float Feature::PixelIrradiance(cv::Point2f pose, cv::Mat image) const
|
||||
{
|
||||
|
||||
return ((float)image.at<uint8_t>(pose.y, pose.x))/255;
|
||||
return ((float)image.at<uint8_t>(pose.y, pose.x))/255.;
|
||||
}
|
||||
|
||||
cv::Point2f Feature::pixelDistanceAt(
|
||||
@ -804,13 +967,18 @@ cv::Point2f Feature::pixelDistanceAt(
|
||||
cam.distortion_coeffs,
|
||||
pure);
|
||||
|
||||
// returns the absolute pixel distance at pixels one metres away
|
||||
// transfrom position to camera frame
|
||||
// to get distance multiplier
|
||||
Eigen::Matrix3d R_w_c0 = quaternionToRotation(cam_state.orientation);
|
||||
const Eigen::Vector3d& t_c0_w = cam_state.position;
|
||||
Eigen::Vector3d p_c0 = R_w_c0 * (in_p-t_c0_w);
|
||||
|
||||
// returns the distance between the pixel points in space
|
||||
cv::Point2f distance(fabs(pure[0].x - pure[1].x), fabs(pure[2].y - pure[3].y));
|
||||
|
||||
return distance;
|
||||
}
|
||||
|
||||
|
||||
cv::Point2f Feature::projectPositionToCamera(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
@ -820,26 +988,40 @@ cv::Point2f Feature::projectPositionToCamera(
|
||||
Eigen::Isometry3d T_c0_w;
|
||||
|
||||
cv::Point2f out_p;
|
||||
|
||||
cv::Point2f my_p;
|
||||
// transfrom position to camera frame
|
||||
|
||||
// cam0 position
|
||||
Eigen::Matrix3d R_w_c0 = quaternionToRotation(cam_state.orientation);
|
||||
const Eigen::Vector3d& t_c0_w = cam_state.position;
|
||||
Eigen::Vector3d p_c0 = R_w_c0 * (in_p-t_c0_w);
|
||||
|
||||
out_p = cv::Point2f(p_c0(0)/p_c0(2), p_c0(1)/p_c0(2));
|
||||
// project point according to model
|
||||
if(cam.id == 0)
|
||||
{
|
||||
Eigen::Vector3d p_c0 = R_w_c0 * (in_p-t_c0_w);
|
||||
out_p = cv::Point2f(p_c0(0)/p_c0(2), p_c0(1)/p_c0(2));
|
||||
|
||||
// if(cam_state_id == observations.begin()->first)
|
||||
//printf("undist:\n \tproj pos: %f, %f\n\ttrue pos: %f, %f\n", out_p.x, out_p.y, undist_anchor_center_pos.x, undist_anchor_center_pos.y);
|
||||
}
|
||||
// if camera is one, calcualte the cam1 position from cam0 position first
|
||||
else if(cam.id == 1)
|
||||
{
|
||||
// cam1 position
|
||||
Eigen::Matrix3d R_c0_c1 = CAMState::T_cam0_cam1.linear();
|
||||
Eigen::Matrix3d R_w_c1 = R_c0_c1 * R_w_c0;
|
||||
Eigen::Vector3d t_c1_w = t_c0_w - R_w_c1.transpose()*CAMState::T_cam0_cam1.translation();
|
||||
|
||||
cv::Point2f my_p = image_handler::distortPoint(out_p,
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs);
|
||||
|
||||
// printf("truPosition: %f, %f, %f\n", position.x(), position.y(), position.z());
|
||||
// printf("camPosition: %f, %f, %f\n", p_c0(0), p_c0(1), p_c0(2));
|
||||
// printf("Photo projection: %f, %f\n", my_p[0].x, my_p[0].y);
|
||||
Eigen::Vector3d p_c1 = R_w_c1 * (in_p-t_c1_w);
|
||||
out_p = cv::Point2f(p_c1(0)/p_c1(2), p_c1(1)/p_c1(2));
|
||||
}
|
||||
|
||||
// undistort point according to camera model
|
||||
if (cam.distortion_model.substr(0,3) == "pre-")
|
||||
my_p = cv::Point2f(out_p.x * cam.intrinsics[0] + cam.intrinsics[2], out_p.y * cam.intrinsics[1] + cam.intrinsics[3]);
|
||||
else
|
||||
my_p = image_handler::distortPoint(out_p,
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs);
|
||||
return my_p;
|
||||
}
|
||||
|
||||
@ -864,45 +1046,75 @@ bool Feature::initializeAnchor(const CameraCalibration& cam, int N)
|
||||
|
||||
auto anchor = observations.begin();
|
||||
if(cam.moving_window.find(anchor->first) == cam.moving_window.end())
|
||||
{
|
||||
return false;
|
||||
|
||||
}
|
||||
cv::Mat anchorImage = cam.moving_window.find(anchor->first)->second.image;
|
||||
cv::Mat anchorImage_deeper;
|
||||
anchorImage.convertTo(anchorImage_deeper,CV_16S);
|
||||
//TODO remove this?
|
||||
|
||||
|
||||
cv::Sobel(anchorImage_deeper, xderImage, -1, 1, 0, 3);
|
||||
cv::Sobel(anchorImage_deeper, yderImage, -1, 0, 1, 3);
|
||||
|
||||
xderImage/=8.;
|
||||
yderImage/=8.;
|
||||
|
||||
cv::convertScaleAbs(xderImage, abs_xderImage);
|
||||
cv::convertScaleAbs(yderImage, abs_yderImage);
|
||||
|
||||
cv::GaussianBlur(anchorImage, anchorImage_blurred, cv::Size(3,3), 0, 0, cv::BORDER_DEFAULT);
|
||||
|
||||
auto u = anchor->second(0);//*cam.intrinsics[0] + cam.intrinsics[2];
|
||||
auto v = anchor->second(1);//*cam.intrinsics[1] + cam.intrinsics[3];
|
||||
|
||||
//testing
|
||||
undist_anchor_center_pos = cv::Point2f(u,v);
|
||||
|
||||
//for NxN patch pixels around feature
|
||||
int count = 0;
|
||||
|
||||
// get feature in undistorted pixel space
|
||||
// this only reverts from 'pure' space into undistorted pixel space using camera matrix
|
||||
cv::Point2f und_pix_p = image_handler::distortPoint(cv::Point2f(u, v),
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs);
|
||||
// create vector of patch in pixel plane
|
||||
for(double u_run = -n; u_run <= n; u_run++)
|
||||
for(double v_run = -n; v_run <= n; v_run++)
|
||||
anchorPatch_real.push_back(cv::Point2f(und_pix_p.x+u_run, und_pix_p.y+v_run));
|
||||
// check if image has been pre-undistorted
|
||||
if(cam.distortion_model.substr(0,3) == "pre")
|
||||
{
|
||||
//project onto pixel plane
|
||||
undist_anchor_center_pos = cv::Point2f(u * cam.intrinsics[0] + cam.intrinsics[2], v * cam.intrinsics[1] + cam.intrinsics[3]);
|
||||
// create vector of patch in pixel plane
|
||||
for(double u_run = -n; u_run <= n; u_run++)
|
||||
for(double v_run = -n; v_run <= n; v_run++)
|
||||
anchorPatch_real.push_back(cv::Point2f(undist_anchor_center_pos.x+u_run, undist_anchor_center_pos.y+v_run));
|
||||
|
||||
|
||||
//create undistorted pure points
|
||||
image_handler::undistortPoints(anchorPatch_real,
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs,
|
||||
anchorPatch_ideal);
|
||||
//project back into u,v
|
||||
for(int i = 0; i < N*N; i++)
|
||||
anchorPatch_ideal.push_back(cv::Point2f((anchorPatch_real[i].x-cam.intrinsics[2])/cam.intrinsics[0], (anchorPatch_real[i].y-cam.intrinsics[3])/cam.intrinsics[1]));
|
||||
}
|
||||
|
||||
else
|
||||
{
|
||||
// get feature in undistorted pixel space
|
||||
// this only reverts from 'pure' space into undistorted pixel space using camera matrix
|
||||
cv::Point2f und_pix_p = image_handler::distortPoint(cv::Point2f(u, v),
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs);
|
||||
// create vector of patch in pixel plane
|
||||
for(double u_run = -n; u_run <= n; u_run++)
|
||||
for(double v_run = -n; v_run <= n; v_run++)
|
||||
anchorPatch_real.push_back(cv::Point2f(und_pix_p.x+u_run, und_pix_p.y+v_run));
|
||||
|
||||
|
||||
//create undistorted pure points
|
||||
image_handler::undistortPoints(anchorPatch_real,
|
||||
cam.intrinsics,
|
||||
cam.distortion_model,
|
||||
cam.distortion_coeffs,
|
||||
anchorPatch_ideal);
|
||||
|
||||
}
|
||||
|
||||
// save anchor position for later visualisaztion
|
||||
anchor_center_pos = anchorPatch_real[(N*N-1)/2];
|
||||
|
||||
// save true pixel Patch position
|
||||
for(auto point : anchorPatch_real)
|
||||
if(point.x - n < 0 || point.x + n >= cam.resolution(0) || point.y - n < 0 || point.y + n >= cam.resolution(1))
|
||||
if(point.x - n < 0 || point.x + n >= cam.resolution(0)-1 || point.y - n < 0 || point.y + n >= cam.resolution(1)-1)
|
||||
return false;
|
||||
|
||||
for(auto point : anchorPatch_real)
|
||||
@ -913,9 +1125,11 @@ bool Feature::initializeAnchor(const CameraCalibration& cam, int N)
|
||||
anchorPatch_3d.push_back(AnchorPixelToPosition(point, cam));
|
||||
|
||||
is_anchored = true;
|
||||
|
||||
return true;
|
||||
}
|
||||
/*
|
||||
|
||||
|
||||
bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
|
||||
// Organize camera poses and feature observations properly.
|
||||
@ -925,9 +1139,6 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
Eigen::aligned_allocator<Eigen::Vector2d> > measurements(0);
|
||||
|
||||
for (auto& m : observations) {
|
||||
// TODO: This should be handled properly. Normally, the
|
||||
// required camera states should all be available in
|
||||
// the input cam_states buffer.
|
||||
auto cam_state_iter = cam_states.find(m.first);
|
||||
if (cam_state_iter == cam_states.end()) continue;
|
||||
|
||||
@ -958,13 +1169,11 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
pose = pose.inverse() * T_c0_w;
|
||||
|
||||
// Generate initial guess
|
||||
Eigen::Vector3d initial_position(0.0, 0.0, 0.0);
|
||||
generateInitialGuess(cam_poses[cam_poses.size()-1], measurements[0],
|
||||
measurements[measurements.size()-1], initial_position);
|
||||
Eigen::Vector3d solution(
|
||||
initial_position(0)/initial_position(2),
|
||||
initial_position(1)/initial_position(2),
|
||||
1.0/initial_position(2));
|
||||
double initial_depth = 0;
|
||||
initial_depth = generateInitialDepth(cam_poses[cam_poses.size()-1], measurements[0],
|
||||
measurements[measurements.size()-1]);
|
||||
|
||||
double solution = 1.0/initial_depth;
|
||||
|
||||
// Apply Levenberg-Marquart method to solve for the 3d position.
|
||||
double lambda = optimization_config.initial_damping;
|
||||
@ -977,21 +1186,21 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
double total_cost = 0.0;
|
||||
for (int i = 0; i < cam_poses.size(); ++i) {
|
||||
double this_cost = 0.0;
|
||||
cost(cam_poses[i], solution, measurements[i], this_cost);
|
||||
Rhocost(cam_poses[i], solution, measurements[0], measurements[i], this_cost);
|
||||
total_cost += this_cost;
|
||||
}
|
||||
|
||||
// Outer loop.
|
||||
do {
|
||||
Eigen::Matrix3d A = Eigen::Matrix3d::Zero();
|
||||
Eigen::Vector3d b = Eigen::Vector3d::Zero();
|
||||
Eigen::Matrix<double, 1, 1> A = Eigen::Matrix<double, 1, 1>::Zero();
|
||||
Eigen::Matrix<double, 1, 1> b = Eigen::Matrix<double, 1, 1>::Zero();
|
||||
|
||||
for (int i = 0; i < cam_poses.size(); ++i) {
|
||||
Eigen::Matrix<double, 2, 3> J;
|
||||
Eigen::Matrix<double, 2, 1> J;
|
||||
Eigen::Vector2d r;
|
||||
double w;
|
||||
|
||||
jacobian(cam_poses[i], solution, measurements[i], J, r, w);
|
||||
RhoJacobian(cam_poses[i], solution, measurements[0], measurements[i], J, r, w);
|
||||
|
||||
if (w == 1) {
|
||||
A += J.transpose() * J;
|
||||
@ -1006,15 +1215,15 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
// Inner loop.
|
||||
// Solve for the delta that can reduce the total cost.
|
||||
do {
|
||||
Eigen::Matrix3d damper = lambda * Eigen::Matrix3d::Identity();
|
||||
Eigen::Vector3d delta = (A+damper).ldlt().solve(b);
|
||||
Eigen::Vector3d new_solution = solution - delta;
|
||||
Eigen::Matrix<double, 1, 1> damper = lambda*Eigen::Matrix<double, 1, 1>::Identity();
|
||||
Eigen::Matrix<double, 1, 1> delta = (A+damper).ldlt().solve(b);
|
||||
double new_solution = solution - delta(0,0);
|
||||
delta_norm = delta.norm();
|
||||
|
||||
double new_cost = 0.0;
|
||||
for (int i = 0; i < cam_poses.size(); ++i) {
|
||||
double this_cost = 0.0;
|
||||
cost(cam_poses[i], new_solution, measurements[i], this_cost);
|
||||
Rhocost(cam_poses[i], new_solution, measurements[0], measurements[i], this_cost);
|
||||
new_cost += this_cost;
|
||||
}
|
||||
|
||||
@ -1039,8 +1248,8 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
|
||||
// Covert the feature position from inverse depth
|
||||
// representation to its 3d coordinate.
|
||||
Eigen::Vector3d final_position(solution(0)/solution(2),
|
||||
solution(1)/solution(2), 1.0/solution(2));
|
||||
Eigen::Vector3d final_position(measurements[0](0)/solution,
|
||||
measurements[0](1)/solution, 1.0/solution);
|
||||
|
||||
// Check if the solution is valid. Make sure the feature
|
||||
// is in front of every camera frame observing it.
|
||||
@ -1055,7 +1264,7 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
}
|
||||
|
||||
//save inverse depth distance from camera
|
||||
anchor_rho = solution(2);
|
||||
anchor_rho = solution;
|
||||
|
||||
// Convert the feature position to the world frame.
|
||||
position = T_c0_w.linear()*final_position + T_c0_w.translation();
|
||||
@ -1065,10 +1274,12 @@ bool Feature::initializeRho(const CamStateServer& cam_states) {
|
||||
|
||||
return is_valid_solution;
|
||||
}
|
||||
*/
|
||||
|
||||
|
||||
bool Feature::initializePosition(const CamStateServer& cam_states) {
|
||||
|
||||
// Organize camera poses and feature observations properly.
|
||||
|
||||
std::vector<Eigen::Isometry3d,
|
||||
Eigen::aligned_allocator<Eigen::Isometry3d> > cam_poses(0);
|
||||
std::vector<Eigen::Vector2d,
|
||||
@ -1206,6 +1417,7 @@ bool Feature::initializePosition(const CamStateServer& cam_states) {
|
||||
|
||||
//save inverse depth distance from camera
|
||||
anchor_rho = solution(2);
|
||||
std::cout << "from feature: " << anchor_rho << std::endl;
|
||||
|
||||
// Convert the feature position to the world frame.
|
||||
position = T_c0_w.linear()*final_position + T_c0_w.translation();
|
||||
|
@ -16,6 +16,16 @@ namespace msckf_vio {
|
||||
*/
|
||||
namespace image_handler {
|
||||
|
||||
cv::Point2f pinholeDownProject(const cv::Point2f& p_in, const cv::Vec4d& intrinsics);
|
||||
cv::Point2f pinholeUpProject(const cv::Point2f& p_in, const cv::Vec4d& intrinsics);
|
||||
|
||||
void undistortImage(
|
||||
cv::InputArray src,
|
||||
cv::OutputArray dst,
|
||||
const std::string& distortion_model,
|
||||
const cv::Vec4d& intrinsics,
|
||||
const cv::Vec4d& distortion_coeffs);
|
||||
|
||||
void undistortPoints(
|
||||
const std::vector<cv::Point2f>& pts_in,
|
||||
const cv::Vec4d& intrinsics,
|
||||
|
@ -320,6 +320,8 @@ private:
|
||||
return;
|
||||
}
|
||||
|
||||
bool STREAMPAUSE;
|
||||
|
||||
// Indicate if this is the first image message.
|
||||
bool is_first_img;
|
||||
|
||||
|
@ -13,13 +13,6 @@
|
||||
|
||||
namespace msckf_vio {
|
||||
|
||||
|
||||
inline double absoluted(double a){
|
||||
if(a>0)
|
||||
return a;
|
||||
else return -a;
|
||||
}
|
||||
|
||||
/*
|
||||
* @brief Create a skew-symmetric matrix from a 3-element vector.
|
||||
* @note Performs the operation:
|
||||
|
@ -202,53 +202,111 @@ class MsckfVio {
|
||||
Eigen::Vector4d& r);
|
||||
// This function computes the Jacobian of all measurements viewed
|
||||
// in the given camera states of this feature.
|
||||
void featureJacobian(const FeatureIDType& feature_id,
|
||||
bool featureJacobian(
|
||||
const FeatureIDType& feature_id,
|
||||
const std::vector<StateIDType>& cam_state_ids,
|
||||
Eigen::MatrixXd& H_x, Eigen::VectorXd& r);
|
||||
|
||||
|
||||
void PhotometricMeasurementJacobian(
|
||||
const StateIDType& cam_state_id,
|
||||
const FeatureIDType& feature_id,
|
||||
Eigen::MatrixXd& H_x,
|
||||
Eigen::MatrixXd& H_y,
|
||||
Eigen::VectorXd& r);
|
||||
void twodotMeasurementJacobian(
|
||||
const StateIDType& cam_state_id,
|
||||
const FeatureIDType& feature_id,
|
||||
Eigen::MatrixXd& H_x, Eigen::MatrixXd& H_y, Eigen::VectorXd& r);
|
||||
|
||||
void PhotometricFeatureJacobian(
|
||||
const FeatureIDType& feature_id,
|
||||
const std::vector<StateIDType>& cam_state_ids,
|
||||
Eigen::MatrixXd& H_x, Eigen::VectorXd& r);
|
||||
bool ConstructJacobians(
|
||||
Eigen::MatrixXd& H_rho,
|
||||
Eigen::MatrixXd& H_pl,
|
||||
Eigen::MatrixXd& H_pA,
|
||||
const Feature& feature,
|
||||
const StateIDType& cam_state_id,
|
||||
Eigen::MatrixXd& H_xl,
|
||||
Eigen::MatrixXd& H_yl);
|
||||
|
||||
bool PhotometricPatchPointResidual(
|
||||
const StateIDType& cam_state_id,
|
||||
const Feature& feature,
|
||||
Eigen::VectorXd& r);
|
||||
|
||||
bool PhotometricPatchPointJacobian(
|
||||
const CAMState& cam_state,
|
||||
const StateIDType& cam_state_id,
|
||||
const Feature& feature,
|
||||
Eigen::Vector3d point,
|
||||
int count,
|
||||
Eigen::Matrix<double, 2, 1>& H_rhoj,
|
||||
Eigen::Matrix<double, 2, 6>& H_plj,
|
||||
Eigen::Matrix<double, 2, 6>& H_pAj,
|
||||
Eigen::Matrix<double, 2, 4>& dI_dhj);
|
||||
|
||||
bool PhotometricMeasurementJacobian(
|
||||
const StateIDType& cam_state_id,
|
||||
const FeatureIDType& feature_id,
|
||||
Eigen::MatrixXd& H_x,
|
||||
Eigen::MatrixXd& H_y,
|
||||
Eigen::VectorXd& r);
|
||||
|
||||
|
||||
bool twodotFeatureJacobian(
|
||||
const FeatureIDType& feature_id,
|
||||
const std::vector<StateIDType>& cam_state_ids,
|
||||
Eigen::MatrixXd& H_x, Eigen::VectorXd& r);
|
||||
|
||||
bool PhotometricFeatureJacobian(
|
||||
const FeatureIDType& feature_id,
|
||||
const std::vector<StateIDType>& cam_state_ids,
|
||||
Eigen::MatrixXd& H_x, Eigen::VectorXd& r);
|
||||
|
||||
void photometricMeasurementUpdate(const Eigen::MatrixXd& H, const Eigen::VectorXd& r);
|
||||
void measurementUpdate(const Eigen::MatrixXd& H,
|
||||
const Eigen::VectorXd& r);
|
||||
void twoMeasurementUpdate(const Eigen::MatrixXd& H, const Eigen::VectorXd& r);
|
||||
|
||||
bool gatingTest(const Eigen::MatrixXd& H,
|
||||
const Eigen::VectorXd&r, const int& dof);
|
||||
const Eigen::VectorXd&r, const int& dof, int filter=0);
|
||||
void removeLostFeatures();
|
||||
void findRedundantCamStates(
|
||||
std::vector<StateIDType>& rm_cam_state_ids);
|
||||
|
||||
void pruneLastCamStateBuffer();
|
||||
void pruneCamStateBuffer();
|
||||
// Reset the system online if the uncertainty is too large.
|
||||
void onlineReset();
|
||||
|
||||
// Photometry flag
|
||||
bool PHOTOMETRIC;
|
||||
int FILTER;
|
||||
|
||||
// debug flag
|
||||
bool STREAMPAUSE;
|
||||
bool PRINTIMAGES;
|
||||
bool GROUNDTRUTH;
|
||||
|
||||
bool nan_flag;
|
||||
bool play;
|
||||
double last_time_bound;
|
||||
double time_offset;
|
||||
|
||||
// Patch size for Photometry
|
||||
int N;
|
||||
|
||||
// Image rescale
|
||||
int SCALE;
|
||||
// Chi squared test table.
|
||||
static std::map<int, double> chi_squared_test_table;
|
||||
|
||||
double eval_time;
|
||||
|
||||
IMUState timed_old_imu_state;
|
||||
IMUState timed_old_true_state;
|
||||
|
||||
IMUState old_imu_state;
|
||||
IMUState old_true_state;
|
||||
|
||||
// change in position
|
||||
Eigen::Vector3d delta_position;
|
||||
Eigen::Vector3d delta_orientation;
|
||||
|
||||
// State vector
|
||||
StateServer state_server;
|
||||
StateServer photometric_state_server;
|
||||
|
||||
// Ground truth state vector
|
||||
StateServer true_state_server;
|
||||
@ -311,6 +369,7 @@ class MsckfVio {
|
||||
// Subscribers and publishers
|
||||
ros::Subscriber imu_sub;
|
||||
ros::Subscriber truth_sub;
|
||||
ros::Publisher truth_odom_pub;
|
||||
ros::Publisher odom_pub;
|
||||
ros::Publisher marker_pub;
|
||||
ros::Publisher feature_pub;
|
||||
|
38
launch/image_processor_tinytum.launch
Normal file
38
launch/image_processor_tinytum.launch
Normal file
@ -0,0 +1,38 @@
|
||||
<launch>
|
||||
|
||||
<arg name="robot" default="firefly_sbx"/>
|
||||
<arg name="calibration_file"
|
||||
default="$(find msckf_vio)/config/camchain-imucam-tum-scaled.yaml"/>
|
||||
|
||||
<!-- Image Processor Nodelet -->
|
||||
<group ns="$(arg robot)">
|
||||
<node pkg="nodelet" type="nodelet" name="image_processor"
|
||||
args="standalone msckf_vio/ImageProcessorNodelet"
|
||||
output="screen"
|
||||
>
|
||||
|
||||
<!-- Debugging Flaggs -->
|
||||
<param name="StreamPause" value="true"/>
|
||||
|
||||
<rosparam command="load" file="$(arg calibration_file)"/>
|
||||
<param name="grid_row" value="4"/>
|
||||
<param name="grid_col" value="4"/>
|
||||
<param name="grid_min_feature_num" value="3"/>
|
||||
<param name="grid_max_feature_num" value="5"/>
|
||||
<param name="pyramid_levels" value="3"/>
|
||||
<param name="patch_size" value="15"/>
|
||||
<param name="fast_threshold" value="10"/>
|
||||
<param name="max_iteration" value="30"/>
|
||||
<param name="track_precision" value="0.01"/>
|
||||
<param name="ransac_threshold" value="3"/>
|
||||
<param name="stereo_threshold" value="5"/>
|
||||
|
||||
<remap from="~imu" to="/imu0"/>
|
||||
<remap from="~cam0_image" to="/cam0/image_raw"/>
|
||||
<remap from="~cam1_image" to="/cam1/image_raw"/>
|
||||
|
||||
|
||||
</node>
|
||||
</group>
|
||||
|
||||
</launch>
|
@ -11,6 +11,9 @@
|
||||
output="screen"
|
||||
>
|
||||
|
||||
<!-- Debugging Flaggs -->
|
||||
<param name="StreamPause" value="true"/>
|
||||
|
||||
<rosparam command="load" file="$(arg calibration_file)"/>
|
||||
<param name="grid_row" value="4"/>
|
||||
<param name="grid_col" value="4"/>
|
||||
|
@ -22,10 +22,10 @@
|
||||
<param name="PHOTOMETRIC" value="true"/>
|
||||
|
||||
<!-- Debugging Flaggs -->
|
||||
<param name="PrintImages" value="false"/>
|
||||
<param name="PrintImages" value="true"/>
|
||||
<param name="GroundTruth" value="false"/>
|
||||
|
||||
<param name="patch_size_n" value="7"/>
|
||||
<param name="patch_size_n" value="3"/>
|
||||
|
||||
<!-- Calibration parameters -->
|
||||
<rosparam command="load" file="$(arg calibration_file)"/>
|
||||
|
@ -17,6 +17,18 @@
|
||||
args='standalone msckf_vio/MsckfVioNodelet'
|
||||
output="screen">
|
||||
|
||||
|
||||
<!-- Filter Flag, 0 = msckf, 1 = photometric, 2 = two -->
|
||||
<param name="FILTER" value="1"/>
|
||||
|
||||
<!-- Debugging Flaggs -->
|
||||
<param name="StreamPause" value="true"/>
|
||||
<param name="PrintImages" value="true"/>
|
||||
<param name="GroundTruth" value="false"/>
|
||||
|
||||
<param name="patch_size_n" value="5"/>
|
||||
<param name="image_scale" value ="1"/>
|
||||
|
||||
<!-- Calibration parameters -->
|
||||
<rosparam command="load" file="$(arg calibration_file)"/>
|
||||
|
||||
|
77
launch/msckf_vio_tinytum.launch
Normal file
77
launch/msckf_vio_tinytum.launch
Normal file
@ -0,0 +1,77 @@
|
||||
<launch>
|
||||
|
||||
<arg name="robot" default="firefly_sbx"/>
|
||||
<arg name="fixed_frame_id" default="world"/>
|
||||
<arg name="calibration_file"
|
||||
default="$(find msckf_vio)/config/camchain-imucam-tum-scaled.yaml"/>
|
||||
|
||||
<!-- Image Processor Nodelet -->
|
||||
<include file="$(find msckf_vio)/launch/image_processor_tinytum.launch">
|
||||
<arg name="robot" value="$(arg robot)"/>
|
||||
<arg name="calibration_file" value="$(arg calibration_file)"/>
|
||||
</include>
|
||||
|
||||
<!-- Msckf Vio Nodelet -->
|
||||
<group ns="$(arg robot)">
|
||||
<node pkg="nodelet" type="nodelet" name="vio"
|
||||
args='standalone msckf_vio/MsckfVioNodelet'
|
||||
output="screen">
|
||||
|
||||
<param name="FILTER" value="0"/>
|
||||
|
||||
<!-- Debugging Flaggs -->
|
||||
<param name="StreamPause" value="true"/>
|
||||
<param name="PrintImages" value="false"/>
|
||||
<param name="GroundTruth" value="false"/>
|
||||
|
||||
<param name="patch_size_n" value="3"/>
|
||||
<param name="image_scale" value ="1"/>
|
||||
|
||||
<!-- Calibration parameters -->
|
||||
<rosparam command="load" file="$(arg calibration_file)"/>
|
||||
|
||||
<param name="publish_tf" value="true"/>
|
||||
<param name="frame_rate" value="20"/>
|
||||
<param name="fixed_frame_id" value="$(arg fixed_frame_id)"/>
|
||||
<param name="child_frame_id" value="odom"/>
|
||||
<param name="max_cam_state_size" value="12"/>
|
||||
<param name="position_std_threshold" value="8.0"/>
|
||||
|
||||
<param name="rotation_threshold" value="0.2618"/>
|
||||
<param name="translation_threshold" value="0.4"/>
|
||||
<param name="tracking_rate_threshold" value="0.5"/>
|
||||
|
||||
<!-- Feature optimization config -->
|
||||
<param name="feature/config/translation_threshold" value="-1.0"/>
|
||||
|
||||
<!-- These values should be standard deviation -->
|
||||
<param name="noise/gyro" value="0.005"/>
|
||||
<param name="noise/acc" value="0.05"/>
|
||||
<param name="noise/gyro_bias" value="0.001"/>
|
||||
<param name="noise/acc_bias" value="0.01"/>
|
||||
<param name="noise/feature" value="0.035"/>
|
||||
|
||||
<param name="initial_state/velocity/x" value="0.0"/>
|
||||
<param name="initial_state/velocity/y" value="0.0"/>
|
||||
<param name="initial_state/velocity/z" value="0.0"/>
|
||||
|
||||
<!-- These values should be covariance -->
|
||||
<param name="initial_covariance/velocity" value="0.25"/>
|
||||
<param name="initial_covariance/gyro_bias" value="0.01"/>
|
||||
<param name="initial_covariance/acc_bias" value="0.01"/>
|
||||
<param name="initial_covariance/extrinsic_rotation_cov" value="3.0462e-4"/>
|
||||
<param name="initial_covariance/extrinsic_translation_cov" value="2.5e-5"/>
|
||||
<param name="initial_covariance/irradiance_frame_bias" value="0.1"/>
|
||||
|
||||
<remap from="~imu" to="/imu0"/>
|
||||
<remap from="~ground_truth" to="/vrpn_client/raw_transform"/>
|
||||
<remap from="~cam0_image" to="/cam0/image_raw"/>
|
||||
<remap from="~cam1_image" to="/cam1/image_raw"/>
|
||||
|
||||
<remap from="~features" to="image_processor/features"/>
|
||||
|
||||
</node>
|
||||
</group>
|
||||
|
||||
|
||||
</launch>
|
@ -17,14 +17,16 @@
|
||||
args='standalone msckf_vio/MsckfVioNodelet'
|
||||
output="screen">
|
||||
|
||||
<!-- Photometry Flag-->
|
||||
<param name="PHOTOMETRIC" value="true"/>
|
||||
|
||||
<!-- Filter Flag, 0 = msckf, 1 = photometric, 2 = two -->
|
||||
<param name="FILTER" value="1"/>
|
||||
|
||||
<!-- Debugging Flaggs -->
|
||||
<param name="PrintImages" value="true"/>
|
||||
<param name="StreamPause" value="true"/>
|
||||
<param name="PrintImages" value="false"/>
|
||||
<param name="GroundTruth" value="false"/>
|
||||
|
||||
<param name="patch_size_n" value="7"/>
|
||||
<param name="patch_size_n" value="3"/>
|
||||
<!-- Calibration parameters -->
|
||||
<rosparam command="load" file="$(arg calibration_file)"/>
|
||||
|
||||
@ -71,4 +73,6 @@
|
||||
</node>
|
||||
</group>
|
||||
|
||||
<!--node name="player" pkg="bagcontrol" type="control.py" /-->
|
||||
|
||||
</launch>
|
||||
|
73
log
Normal file
73
log
Normal file
@ -0,0 +1,73 @@
|
||||
# Created by Octave 3.8.1, Wed Jun 12 14:36:37 2019 CEST <raphael@raphael-desktop>
|
||||
# name: Hx
|
||||
# type: matrix
|
||||
# rows: 18
|
||||
# columns: 49
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 313.795 -58.7912 139.778 -46.7616 144.055 86.9644 0 -314.123 55.6434 -140.648 46.7616 -144.055 -86.9644 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 441.06 -94.50069999999999 174.424 -53.7653 204.822 120.248 0 -441.685 90.1101 -175.657 53.7653 -204.822 -120.248 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 225.35 -54.5629 77.60599999999999 -21.1425 105.886 60.3706 0 -225.756 52.3373 -78.2406 21.1425 -105.886 -60.3706 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 175.128 20.6203 175.127 -79.63939999999999 73.245 62.1868 0 -174.573 -22.5235 -175.576 79.63939999999999 -73.245 -62.1868 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 296.962 43.5469 311.307 -143.667 123.399 108.355 0 -295.905 -46.7952 -312.063 143.667 -123.399 -108.355 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 126.283 117.889 311.864 -161.264 38.8521 71.8019 0 -124.464 -119.541 -312.118 161.264 -38.8521 -71.8019 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 49.2502 63.7166 155.071 -80.81950000000001 12.7732 32.1826 0 -48.2934 -64.4113 -155.157 80.81950000000001 -12.7732 -32.1826 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 69.59699999999999 154.579 335.384 -179.355 9.212580000000001 62.0364 0 -67.35599999999999 -155.735 -335.462 179.355 -9.212580000000001 -62.0364 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -66.6965 304.947 500.218 -285.589 -71.31010000000001 55.5058 0 70.8009 -305.077 -499.831 285.589 71.31010000000001 -55.5058 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 323.404 -62.2043 141.092 -46.6015 148.737 89.1108 0 0 0 0 0 0 0 0 -324.336 57.8552 -141.991 46.6015 -148.737 -89.1108 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 454.208 -99.3095 175.986 -53.4094 211.276 123.174 0 0 0 0 0 0 0 0 -455.779 93.0992 -177.158 53.4094 -211.276 -123.174 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 231.884 -57.0266 78.2559 -20.8926 109.118 61.8183 0 0 0 0 0 0 0 0 -232.824 53.8025 -78.80719999999999 20.8926 -109.118 -61.8183 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 181.715 18.8525 177.045 -80.08499999999999 76.3716 63.8254 0 0 0 0 0 0 0 0 -181.07 -20.839 -177.959 80.08499999999999 -76.3716 -63.8254 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 308.249 40.5812 314.711 -144.494 128.766 111.207 0 0 0 0 0 0 0 0 -306.972 -43.8825 -316.328 144.494 -128.766 -111.207 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 133.309 116.865 315.474 -162.598 42.0763 73.8353 0 0 0 0 0 0 0 0 -130.603 -117.454 -316.931 162.598 -42.0763 -73.8353 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 52.4426 63.3607 156.902 -81.5288 14.2139 33.1222 0 0 0 0 0 0 0 0 -50.9976 -63.4393 -157.607 81.5288 -14.2139 -33.1222 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 75.5508 154.234 339.369 -180.995 11.859 63.8956 0 0 0 0 0 0 0 0 -72.1041 -153.816 -340.865 180.995 -11.859 -63.8956 1 0 0 0 0 0 0 0
|
||||
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -62.0551 306.357 506.351 -288.542 -69.50409999999999 57.4562 0 0 0 0 0 0 0 0 68.6262 -303.028 -508.423 288.542 69.50409999999999 -57.4562 1 0 0 0 0 0 0 0
|
||||
|
||||
|
||||
# name: Hy
|
||||
# type: matrix
|
||||
# rows: 18
|
||||
# columns: 14
|
||||
1.56267 0 0 0 0 0 0 0 0 0 0.252873 0 0 0.110387
|
||||
0 1.56267 0 0 0 0 0 0 0 0 0.453168 0 0 0.162961
|
||||
0 0 1.56267 0 0 0 0 0 0 0 0.638441 0 0 0.0873758
|
||||
0 0 0 1.56267 0 0 0 0 0 0 0.21031 0 0 0.0321517
|
||||
0 0 0 0 1.56267 0 0 0 0 0 0.375554 0 0 0.0505151
|
||||
0 0 0 0 0 1.56267 0 0 0 0 0.638441 0 0 -0.0336034
|
||||
0 0 0 0 0 0 1.56267 0 0 0 0.157733 0 0 -0.0232704
|
||||
0 0 0 0 0 0 0 1.56267 0 0 0.212814 0 0 -0.0688343
|
||||
0 0 0 0 0 0 0 0 1.56267 0 0.360532 0 0 -0.187745
|
||||
1.56267 0 0 0 0 0 0 0 0 0 0 0.252873 0 0.322811
|
||||
0 1.56267 0 0 0 0 0 0 0 0 0 0.453168 0 0.467319
|
||||
0 0 1.56267 0 0 0 0 0 0 0 0 0.638441 0 0.245907
|
||||
0 0 0 1.56267 0 0 0 0 0 0 0 0.21031 0 0.135725
|
||||
0 0 0 0 1.56267 0 0 0 0 0 0 0.375554 0 0.225277
|
||||
0 0 0 0 0 1.56267 0 0 0 0 0 0.638441 0 0.012926
|
||||
0 0 0 0 0 0 1.56267 0 0 0 0 0.157733 0 -0.0108962
|
||||
0 0 0 0 0 0 0 1.56267 0 0 0 0.212814 0 -0.06974909999999999
|
||||
0 0 0 0 0 0 0 0 1.56267 0 0 0.360532 0 -0.318846
|
||||
|
||||
|
||||
# name: r
|
||||
# type: matrix
|
||||
# rows: 18
|
||||
# columns: 1
|
||||
0.0354809
|
||||
0.0153183
|
||||
0.0570191
|
||||
-0.0372801
|
||||
0.0878601
|
||||
0.06811780000000001
|
||||
-0.00426164
|
||||
5.162985999041026e-321
|
||||
6.927779999999998e-310
|
||||
0
|
||||
2.121999999910509e-314
|
||||
0
|
||||
0
|
||||
0
|
||||
3.6073900000086e-313
|
||||
4.446590812571219e-323
|
||||
3.952525166729972e-323
|
||||
3.952525166729972e-323
|
||||
|
||||
|
64
src/bagcontrol.py
Normal file
64
src/bagcontrol.py
Normal file
@ -0,0 +1,64 @@
|
||||
#!/usr/bin/env python
|
||||
import rosbag
|
||||
import rospy
|
||||
from sensor_msgs.msg import Imu, Image
|
||||
from geometry_msgs.msg import TransformStamped
|
||||
import time
|
||||
import signal
|
||||
import sys
|
||||
|
||||
|
||||
def signal_handler(sig, frame):
|
||||
print('gracefully exiting the program.')
|
||||
bag.close()
|
||||
sys.exit(0)
|
||||
|
||||
def main():
|
||||
global bag
|
||||
|
||||
cam0_topic = '/cam0/image_raw'
|
||||
cam1_topic = '/cam1/image_raw'
|
||||
imu0_topic = '/imu0'
|
||||
grnd_topic = '/vrpn_client/raw_transform'
|
||||
|
||||
rospy.init_node('controlbag')
|
||||
rospy.set_param('play_bag', False)
|
||||
|
||||
cam0_pub = rospy.Publisher(cam0_topic, Image, queue_size=10)
|
||||
cam1_pub = rospy.Publisher(cam1_topic, Image, queue_size=10)
|
||||
imu0_pub = rospy.Publisher(imu0_topic, Imu, queue_size=10)
|
||||
grnd_pub = rospy.Publisher(grnd_topic, TransformStamped, queue_size=10)
|
||||
|
||||
signal.signal(signal.SIGINT, signal_handler)
|
||||
|
||||
bag = rosbag.Bag('/home/raphael/dev/MSCKF_ws/bag/TUM/dataset-corridor1_1024_16.bag')
|
||||
for topic, msg, t in bag.read_messages(topics=[cam0_topic, cam1_topic, imu0_topic, grnd_topic]):
|
||||
|
||||
# pause if parameter set to false
|
||||
flag = False
|
||||
while not rospy.get_param('/play_bag'):
|
||||
time.sleep(0.01)
|
||||
if not flag:
|
||||
print("stopped playback")
|
||||
flag = not flag
|
||||
|
||||
if flag:
|
||||
print("resume playback")
|
||||
|
||||
if topic == cam0_topic:
|
||||
cam0_pub.publish(msg)
|
||||
|
||||
elif topic == cam1_topic:
|
||||
cam1_pub.publish(msg)
|
||||
|
||||
elif topic == imu0_topic:
|
||||
imu0_pub.publish(msg)
|
||||
|
||||
elif topic ==grnd_topic:
|
||||
grnd_pub.publish(msg)
|
||||
|
||||
#print msg
|
||||
bag.close()
|
||||
|
||||
if __name__== "__main__":
|
||||
main()
|
64
src/control.py
Executable file
64
src/control.py
Executable file
@ -0,0 +1,64 @@
|
||||
#!/usr/bin/env python
|
||||
import rosbag
|
||||
import rospy
|
||||
from sensor_msgs.msg import Imu, Image
|
||||
from geometry_msgs.msg import TransformStamped
|
||||
import time
|
||||
import signal
|
||||
import sys
|
||||
|
||||
|
||||
def signal_handler(sig, frame):
|
||||
print('gracefully exiting the program.')
|
||||
bag.close()
|
||||
sys.exit(0)
|
||||
|
||||
def main():
|
||||
global bag
|
||||
|
||||
cam0_topic = '/cam0/image_raw'
|
||||
cam1_topic = '/cam1/image_raw'
|
||||
imu0_topic = '/imu0'
|
||||
grnd_topic = '/vrpn_client/raw_transform'
|
||||
|
||||
rospy.init_node('controlbag')
|
||||
rospy.set_param('play_bag', False)
|
||||
|
||||
cam0_pub = rospy.Publisher(cam0_topic, Image, queue_size=10)
|
||||
cam1_pub = rospy.Publisher(cam1_topic, Image, queue_size=10)
|
||||
imu0_pub = rospy.Publisher(imu0_topic, Imu, queue_size=10)
|
||||
grnd_pub = rospy.Publisher(grnd_topic, TransformStamped, queue_size=10)
|
||||
|
||||
signal.signal(signal.SIGINT, signal_handler)
|
||||
|
||||
bag = rosbag.Bag('/home/raphael/dev/MSCKF_ws/bag/TUM/dataset-corridor1_1024_16.bag')
|
||||
for topic, msg, t in bag.read_messages(topics=[cam0_topic, cam1_topic, imu0_topic, grnd_topic]):
|
||||
|
||||
# pause if parameter set to false
|
||||
flag = False
|
||||
while not rospy.get_param('/play_bag'):
|
||||
time.sleep(0.01)
|
||||
if not flag:
|
||||
print("stopped playback")
|
||||
flag = not flag
|
||||
|
||||
if flag:
|
||||
print("resume playback")
|
||||
|
||||
if topic == cam0_topic:
|
||||
cam0_pub.publish(msg)
|
||||
|
||||
elif topic == cam1_topic:
|
||||
cam1_pub.publish(msg)
|
||||
|
||||
elif topic == imu0_topic:
|
||||
imu0_pub.publish(msg)
|
||||
|
||||
elif topic ==grnd_topic:
|
||||
grnd_pub.publish(msg)
|
||||
|
||||
#print msg
|
||||
bag.close()
|
||||
|
||||
if __name__== "__main__":
|
||||
main()
|
@ -14,6 +14,40 @@ namespace msckf_vio {
|
||||
namespace image_handler {
|
||||
|
||||
|
||||
cv::Point2f pinholeDownProject(const cv::Point2f& p_in, const cv::Vec4d& intrinsics)
|
||||
{
|
||||
return cv::Point2f(p_in.x * intrinsics[0] + intrinsics[2], p_in.y * intrinsics[1] + intrinsics[3]);
|
||||
}
|
||||
|
||||
cv::Point2f pinholeUpProject(const cv::Point2f& p_in, const cv::Vec4d& intrinsics)
|
||||
{
|
||||
return cv::Point2f((p_in.x - intrinsics[2])/intrinsics[0], (p_in.y - intrinsics[3])/intrinsics[1]);
|
||||
}
|
||||
|
||||
void undistortImage(
|
||||
cv::InputArray src,
|
||||
cv::OutputArray dst,
|
||||
const std::string& distortion_model,
|
||||
const cv::Vec4d& intrinsics,
|
||||
const cv::Vec4d& distortion_coeffs)
|
||||
{
|
||||
|
||||
const cv::Matx33d K(intrinsics[0], 0.0, intrinsics[2],
|
||||
0.0, intrinsics[1], intrinsics[3],
|
||||
0.0, 0.0, 1.0);
|
||||
|
||||
if (distortion_model == "pre-equidistant")
|
||||
cv::fisheye::undistortImage(src, dst, K, distortion_coeffs, K);
|
||||
else if (distortion_model == "equidistant")
|
||||
src.copyTo(dst);
|
||||
|
||||
else if (distortion_model == "pre-radtan")
|
||||
cv::undistort(src, dst, K, distortion_coeffs, K);
|
||||
else if (distortion_model == "radtan")
|
||||
src.copyTo(dst);
|
||||
|
||||
}
|
||||
|
||||
void undistortPoint(
|
||||
const cv::Point2f& pt_in,
|
||||
const cv::Vec4d& intrinsics,
|
||||
@ -42,10 +76,37 @@ void undistortPoint(
|
||||
if (distortion_model == "radtan") {
|
||||
cv::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
|
||||
rectification_matrix, K_new);
|
||||
} else if (distortion_model == "equidistant") {
|
||||
}
|
||||
// equidistant
|
||||
else if (distortion_model == "equidistant") {
|
||||
cv::fisheye::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
|
||||
rectification_matrix, K_new);
|
||||
} else {
|
||||
}
|
||||
// fov
|
||||
else if (distortion_model == "fov") {
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
{
|
||||
float omega = distortion_coeffs[0];
|
||||
float rd = sqrt(pts_in[i].x * pts_in[i].x + pts_in[i].y * pts_in[i].y);
|
||||
float ru = tan(rd * omega)/(2 * tan(omega / 2));
|
||||
|
||||
cv::Point2f newPoint(
|
||||
((pts_in[i].x - intrinsics[2]) / intrinsics[0]) * (ru / rd),
|
||||
((pts_in[i].y - intrinsics[3]) / intrinsics[1]) * (ru / rd));
|
||||
|
||||
pts_out.push_back(newPoint);
|
||||
}
|
||||
}
|
||||
|
||||
else if (distortion_model == "pre-equidistant" or distortion_model == "pre-radtan")
|
||||
{
|
||||
std::vector<cv::Point2f> temp_pts_out;
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
temp_pts_out.push_back(pinholeUpProject(pts_in[i], intrinsics));
|
||||
|
||||
pts_out = temp_pts_out;
|
||||
}
|
||||
else {
|
||||
ROS_WARN_ONCE("The model %s is unrecognized, use radtan instead...",
|
||||
distortion_model.c_str());
|
||||
cv::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
|
||||
@ -79,10 +140,35 @@ void undistortPoints(
|
||||
if (distortion_model == "radtan") {
|
||||
cv::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
|
||||
rectification_matrix, K_new);
|
||||
} else if (distortion_model == "equidistant") {
|
||||
}
|
||||
else if (distortion_model == "equidistant") {
|
||||
cv::fisheye::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
|
||||
rectification_matrix, K_new);
|
||||
} else {
|
||||
}
|
||||
else if (distortion_model == "fov") {
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
{
|
||||
float omega = distortion_coeffs[0];
|
||||
float rd = sqrt(pts_in[i].x * pts_in[i].x + pts_in[i].y * pts_in[i].y);
|
||||
float ru = tan(rd * omega)/(2 * tan(omega / 2));
|
||||
|
||||
cv::Point2f newPoint(
|
||||
((pts_in[i].x - intrinsics[2]) / intrinsics[0]) * (ru / rd),
|
||||
((pts_in[i].y - intrinsics[3]) / intrinsics[1]) * (ru / rd));
|
||||
|
||||
pts_out.push_back(newPoint);
|
||||
}
|
||||
}
|
||||
else if (distortion_model == "pre-equidistant" or distortion_model == "pre-radtan")
|
||||
{
|
||||
std::vector<cv::Point2f> temp_pts_out;
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
temp_pts_out.push_back(pinholeUpProject(pts_in[i], intrinsics));
|
||||
|
||||
pts_out = temp_pts_out;
|
||||
|
||||
}
|
||||
else {
|
||||
ROS_WARN_ONCE("The model %s is unrecognized, use radtan instead...",
|
||||
distortion_model.c_str());
|
||||
cv::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
|
||||
@ -110,7 +196,31 @@ std::vector<cv::Point2f> distortPoints(
|
||||
distortion_coeffs, pts_out);
|
||||
} else if (distortion_model == "equidistant") {
|
||||
cv::fisheye::distortPoints(pts_in, pts_out, K, distortion_coeffs);
|
||||
} else {
|
||||
}
|
||||
else if (distortion_model == "fov") {
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
{
|
||||
// based on 'straight lines have to be straight'
|
||||
float ru = sqrt(pts_in[i].x * pts_in[i].x + pts_in[i].y * pts_in[i].y);
|
||||
float omega = distortion_coeffs[0];
|
||||
float rd = 1 / (omega)*atan(2*ru*tan(omega / 2));
|
||||
|
||||
cv::Point2f newPoint(
|
||||
pts_in[i].x * (rd/ru) * intrinsics[0] + intrinsics[2],
|
||||
pts_in[i].y * (rd/ru) * intrinsics[1] + intrinsics[3]);
|
||||
|
||||
pts_out.push_back(newPoint);
|
||||
}
|
||||
}
|
||||
else if (distortion_model == "pre-equidistant" or distortion_model == "pre-radtan")
|
||||
{
|
||||
std::vector<cv::Point2f> temp_pts_out;
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
temp_pts_out.push_back(pinholeDownProject(pts_in[i], intrinsics));
|
||||
|
||||
pts_out = temp_pts_out;
|
||||
}
|
||||
else {
|
||||
ROS_WARN_ONCE("The model %s is unrecognized, using radtan instead...",
|
||||
distortion_model.c_str());
|
||||
std::vector<cv::Point3f> homogenous_pts;
|
||||
@ -143,7 +253,31 @@ cv::Point2f distortPoint(
|
||||
distortion_coeffs, pts_out);
|
||||
} else if (distortion_model == "equidistant") {
|
||||
cv::fisheye::distortPoints(pts_in, pts_out, K, distortion_coeffs);
|
||||
} else {
|
||||
}
|
||||
else if (distortion_model == "fov") {
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
{
|
||||
// based on 'straight lines have to be straight'
|
||||
float ru = sqrt(pts_in[i].x * pts_in[i].x + pts_in[i].y * pts_in[i].y);
|
||||
float omega = distortion_coeffs[0];
|
||||
float rd = 1 / (omega)*atan(2*ru*tan(omega / 2));
|
||||
|
||||
cv::Point2f newPoint(
|
||||
pts_in[i].x * (rd/ru) * intrinsics[0] + intrinsics[2],
|
||||
pts_in[i].y * (rd/ru) * intrinsics[1] + intrinsics[3]);
|
||||
|
||||
pts_out.push_back(newPoint);
|
||||
}
|
||||
}
|
||||
else if (distortion_model == "pre-equidistant" or distortion_model == "pre-radtan")
|
||||
{
|
||||
std::vector<cv::Point2f> temp_pts_out;
|
||||
for(int i = 0; i < pts_in.size(); i++)
|
||||
pts_out.push_back(pinholeDownProject(pts_in[i], intrinsics));
|
||||
|
||||
pts_out = temp_pts_out;
|
||||
}
|
||||
else {
|
||||
ROS_WARN_ONCE("The model %s is unrecognized, using radtan instead...",
|
||||
distortion_model.c_str());
|
||||
std::vector<cv::Point3f> homogenous_pts;
|
||||
@ -155,5 +289,5 @@ cv::Point2f distortPoint(
|
||||
return pts_out[0];
|
||||
}
|
||||
|
||||
} // namespace image_handler
|
||||
} // namespace image_handler
|
||||
} // namespace msckf_vio
|
@ -42,6 +42,9 @@ ImageProcessor::~ImageProcessor() {
|
||||
}
|
||||
|
||||
bool ImageProcessor::loadParameters() {
|
||||
|
||||
// debug parameters
|
||||
nh.param<bool>("StreamPause", STREAMPAUSE, false);
|
||||
// Camera calibration parameters
|
||||
nh.param<string>("cam0/distortion_model",
|
||||
cam0.distortion_model, string("radtan"));
|
||||
@ -211,7 +214,9 @@ void ImageProcessor::stereoCallback(
|
||||
const sensor_msgs::ImageConstPtr& cam0_img,
|
||||
const sensor_msgs::ImageConstPtr& cam1_img) {
|
||||
|
||||
//cout << "==================================" << endl;
|
||||
// stop playing bagfile if printing images
|
||||
//if(STREAMPAUSE)
|
||||
// nh.setParam("/play_bag_image", false);
|
||||
|
||||
// Get the current image.
|
||||
cam0_curr_img_ptr = cv_bridge::toCvShare(cam0_img,
|
||||
@ -219,12 +224,27 @@ void ImageProcessor::stereoCallback(
|
||||
cam1_curr_img_ptr = cv_bridge::toCvShare(cam1_img,
|
||||
sensor_msgs::image_encodings::MONO8);
|
||||
|
||||
ros::Time start_time = ros::Time::now();
|
||||
|
||||
|
||||
cv::Mat new_cam0;
|
||||
cv::Mat new_cam1;
|
||||
|
||||
image_handler::undistortImage(cam0_curr_img_ptr->image, new_cam0, cam0.distortion_model, cam0.intrinsics, cam0.distortion_coeffs);
|
||||
image_handler::undistortImage(cam1_curr_img_ptr->image, new_cam1, cam1.distortion_model, cam1.intrinsics, cam1.distortion_coeffs);
|
||||
|
||||
new_cam0.copyTo(cam0_curr_img_ptr->image);
|
||||
new_cam1.copyTo(cam1_curr_img_ptr->image);
|
||||
|
||||
//ROS_INFO("Publishing: %f",
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
|
||||
// Build the image pyramids once since they're used at multiple places
|
||||
createImagePyramids();
|
||||
|
||||
// Detect features in the first frame.
|
||||
if (is_first_img) {
|
||||
ros::Time start_time = ros::Time::now();
|
||||
start_time = ros::Time::now();
|
||||
initializeFirstFrame();
|
||||
//ROS_INFO("Detection time: %f",
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
@ -237,7 +257,7 @@ void ImageProcessor::stereoCallback(
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
} else {
|
||||
// Track the feature in the previous image.
|
||||
ros::Time start_time = ros::Time::now();
|
||||
start_time = ros::Time::now();
|
||||
trackFeatures();
|
||||
//ROS_INFO("Tracking time: %f",
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
@ -245,6 +265,7 @@ void ImageProcessor::stereoCallback(
|
||||
// Add new features into the current image.
|
||||
start_time = ros::Time::now();
|
||||
addNewFeatures();
|
||||
|
||||
//ROS_INFO("Addition time: %f",
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
|
||||
@ -267,16 +288,18 @@ void ImageProcessor::stereoCallback(
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
|
||||
// Publish features in the current image.
|
||||
ros::Time start_time = ros::Time::now();
|
||||
start_time = ros::Time::now();
|
||||
publish();
|
||||
//ROS_INFO("Publishing: %f",
|
||||
// (ros::Time::now()-start_time).toSec());
|
||||
|
||||
// Update the previous image and previous features.
|
||||
|
||||
cam0_prev_img_ptr = cam0_curr_img_ptr;
|
||||
prev_features_ptr = curr_features_ptr;
|
||||
std::swap(prev_cam0_pyramid_, curr_cam0_pyramid_);
|
||||
|
||||
|
||||
// Initialize the current features to empty vectors.
|
||||
curr_features_ptr.reset(new GridFeatures());
|
||||
for (int code = 0; code <
|
||||
@ -284,6 +307,10 @@ void ImageProcessor::stereoCallback(
|
||||
(*curr_features_ptr)[code] = vector<FeatureMetaData>(0);
|
||||
}
|
||||
|
||||
// stop playing bagfile if printing images
|
||||
//if(STREAMPAUSE)
|
||||
// nh.setParam("/play_bag_image", true);
|
||||
|
||||
return;
|
||||
}
|
||||
|
||||
@ -580,6 +607,7 @@ void ImageProcessor::trackFeatures() {
|
||||
++after_ransac;
|
||||
}
|
||||
|
||||
|
||||
// Compute the tracking rate.
|
||||
int prev_feature_num = 0;
|
||||
for (const auto& item : *prev_features_ptr)
|
||||
@ -619,9 +647,14 @@ void ImageProcessor::stereoMatch(
|
||||
image_handler::undistortPoints(cam0_points, cam0.intrinsics, cam0.distortion_model,
|
||||
cam0.distortion_coeffs, cam0_points_undistorted,
|
||||
R_cam0_cam1);
|
||||
<<<<<<< HEAD
|
||||
cam1_points = distortPoints(cam0_points_undistorted, cam1_intrinsics,
|
||||
cam1_distortion_model, cam1_distortion_coeffs);
|
||||
=======
|
||||
|
||||
cam1_points = image_handler::distortPoints(cam0_points_undistorted, cam1.intrinsics,
|
||||
cam1.distortion_model, cam1.distortion_coeffs);
|
||||
>>>>>>> photometry-jakobi
|
||||
}
|
||||
|
||||
// Track features using LK optical flow method.
|
||||
@ -659,6 +692,8 @@ void ImageProcessor::stereoMatch(
|
||||
|
||||
// Further remove outliers based on the known
|
||||
// essential matrix.
|
||||
|
||||
|
||||
vector<cv::Point2f> cam0_points_undistorted(0);
|
||||
vector<cv::Point2f> cam1_points_undistorted(0);
|
||||
image_handler::undistortPoints(
|
||||
@ -668,6 +703,7 @@ void ImageProcessor::stereoMatch(
|
||||
cam1_points, cam1.intrinsics, cam1.distortion_model,
|
||||
cam1.distortion_coeffs, cam1_points_undistorted);
|
||||
|
||||
|
||||
double norm_pixel_unit = 4.0 / (
|
||||
cam0.intrinsics[0]+cam0.intrinsics[1]+
|
||||
cam1.intrinsics[0]+cam1.intrinsics[1]);
|
||||
|
1874
src/msckf_vio.cpp
1874
src/msckf_vio.cpp
File diff suppressed because it is too large
Load Diff
75
src/shrinkImage.py
Executable file
75
src/shrinkImage.py
Executable file
@ -0,0 +1,75 @@
|
||||
#!/usr/bin/env python
|
||||
from __future__ import print_function
|
||||
|
||||
import sys
|
||||
import rospy
|
||||
import cv2
|
||||
from std_msgs.msg import String
|
||||
from sensor_msgs.msg import Image
|
||||
from cv_bridge import CvBridge, CvBridgeError
|
||||
|
||||
class image_converter:
|
||||
|
||||
def __init__(self):
|
||||
self.image0_pub = rospy.Publisher("/cam0/new_image_raw",Image, queue_size=10)
|
||||
self.image1_pub = rospy.Publisher("/cam1/new_image_raw",Image, queue_size=10)
|
||||
|
||||
self.bridge = CvBridge()
|
||||
self.image0_sub = rospy.Subscriber("/cam0/image_raw",Image,self.callback_cam0)
|
||||
self.image1_sub = rospy.Subscriber("/cam1/image_raw",Image,self.callback_cam1)
|
||||
|
||||
def callback_cam0(self,data):
|
||||
try:
|
||||
cv_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
|
||||
except CvBridgeError as e:
|
||||
print(e)
|
||||
|
||||
imgScale = 0.25
|
||||
newX,newY = cv_image.shape[1]*imgScale, cv_image.shape[0]*imgScale
|
||||
newimg = cv2.resize(cv_image,(int(newX),int(newY)))
|
||||
|
||||
newpub = self.bridge.cv2_to_imgmsg(newimg, "bgr8")
|
||||
newdata = data
|
||||
newdata.height = newpub.height
|
||||
newdata.width = newpub.width
|
||||
newdata.step = newpub.step
|
||||
newdata.data = newpub.data
|
||||
try:
|
||||
self.image0_pub.publish(newdata)
|
||||
except CvBridgeError as e:
|
||||
print(e)
|
||||
|
||||
def callback_cam1(self,data):
|
||||
try:
|
||||
cv_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
|
||||
except CvBridgeError as e:
|
||||
print(e)
|
||||
|
||||
imgScale = 0.25
|
||||
newX,newY = cv_image.shape[1]*imgScale, cv_image.shape[0]*imgScale
|
||||
newimg = cv2.resize(cv_image,(int(newX),int(newY)))
|
||||
|
||||
newpub = self.bridge.cv2_to_imgmsg(newimg, "bgr8")
|
||||
newdata = data
|
||||
newdata.height = newpub.height
|
||||
newdata.width = newpub.width
|
||||
newdata.step = newpub.step
|
||||
newdata.data = newpub.data
|
||||
|
||||
try:
|
||||
self.image1_pub.publish(newdata)
|
||||
except CvBridgeError as e:
|
||||
print(e)
|
||||
|
||||
|
||||
def main(args):
|
||||
ic = image_converter()
|
||||
rospy.init_node('image_converter', anonymous=True)
|
||||
try:
|
||||
rospy.spin()
|
||||
except KeyboardInterrupt:
|
||||
print("Shutting down")
|
||||
cv2.destroyAllWindows()
|
||||
|
||||
if __name__ == '__main__':
|
||||
main(sys.argv)
|
Loading…
Reference in New Issue
Block a user