roger-chuh / vins-fisheye-cubemap Goto Github PK
View Code? Open in Web Editor NEWa VINS algorithm with a combination of stereo fisheye images, cubemap, line features, dense mapping and loop closure
License: GNU General Public License v3.0
a VINS algorithm with a combination of stereo fisheye images, cubemap, line features, dense mapping and loop closure
License: GNU General Public License v3.0
Hi @Roger-Chuh Thanks for your sharing code.
I met some problem when I use the code:
assigning pinhle params... , focal_x = 0.000
disable 0 4 5 9: 0 0 0 0
disable 0 4 5 9: 0 0 0 0
disable 0 4 5 9: 0 0 0 0
disable 0 4 5 9: 0 0 0 0
disable 0 4 5 9: 0 0 0 0
disable 0 4 5 9: 0 0 0 0
lsd.size: 62
up_top_img size: 800 x 800
down_top_img size: 800 x 800
up_side_img size: 1426 x 1426
down_side_img size: 1426 x 1426
terminate called after throwing an instance of 'cv::Exception'
what(): OpenCV(3.4.16) /home/rui/gwm/opencv-3.4.16/modules/core/src/matrix.cpp:751: error: (-215:Assertion failed) 0 <= roi.x && 0 <= roi.width && roi.x + roi.width <= m.cols && 0 <= roi.y && 0 <= roi.height && roi.y + roi.height <= m.rows in function 'Mat'
terminate called recursively
environments:
ubuntu16.04, ros kinetic, opencv 3.4.16 cuda with contrib
config:
%YAML:1.0
#common parameters
#support: 1 imu 1 cam; 1 imu 2 cam: 2 cam;
imu: 1
num_of_cam: 2
is_fisheye: 1
imu_topic: "/imu0"
image0_topic: "/cam0/image_raw"
image1_topic: "/cam1/image_raw"
output_path: "/home/rui/output"
depth_config: "depth_cuda.yaml"
cam0_calib: "up.yaml"
cam1_calib: "down.yaml"
image_width: 800 # For fisheye, this indicate the flattened image width; min 100; 300 - 500 is good for vins
image_height: 500
show_width: 2000
#fisheye_fov: 200
lk_pyr_level: 3
lk_win_size: 21
keyframe_longtrack_thres: 20
#debug_image: 0
print_log: 0
long_track_ratio: 0.5
equalize: 0
use_line: 1 # 1 # 0 # 1 # 0 # 1
debug_image: 0
image_height_raw: 2880
image_width_raw: 2880
base_line: -1
show_line: 0 # 0 # 1 # 0 # 1 # 0 # 1 # 0
show_disp: 0
line_angle_thres: 0.35 # 1.0 # 0.35 # 1.0 # 0.35 # 0.5
line_pixel_thres: 2.0 # 5.0 # 2.0 # 13.0 # 2.0 # 1.5
use_multi_line_triangulation: 0 # 1 # 0 # 1 # 0 # 1
min_trace_to_marg: 1 # -5
cube_map: 1
use_new: 1
fisheye_fov_actual: 200
fisheye_fov: 200
center_fov: 100
enable_up_top: 1
enable_down_top: 1
enable_up_side: 1
enable_down_side: 1
enable_rear_side: 1
thres_outlier : 3.0
tri_max_err: 5.0
estimate_extrinsic: 0 # 0 Have an accurate extrinsic parameters. We will trust the following imu^R_cam, imu^T_cam, don't
body_T_cam1: !!opencv-matrix
rows: 4
cols: 4
dt: d
data: [ 0.99990724, -0.00548477,
0.01246729, 0.00548184,
-0.01247326, -0.0010551,
0.99992165, 0.01221265,
-0.00547119, -0.9999844,
-0.00112341, -0.01776372, 0., 0., 0., 1. ]
body_T_cam0: !!opencv-matrix
rows: 4
cols: 4
dt: d
data: [ -0.99996418, 0.00643933,
-0.00549328, 0.00395783,
0.00547534, -0.00280139,
-0.99998109, -0.0081369,
-0.0064546, -0.99997534,
0.00276603, -0.01790324, 0., 0., 0., 1. ]
pub_flatten: 1
flatten_color: 0
warn_imu_freq: 0
imu_freq: 500
image_freq: 24
multiple_thread: 1
#Gpu accleration support
use_vxworks: 0
use_gpu: 0
enable_depth: 0 # If estimate depth cloud; only available for dual fisheye now
rgb_depth_cloud: -1 # -1: point no texture, 0 depth cloud will be gray, 1 depth cloud will be colored;
#Note that textured and colored depth cloud will slow down whole system
depth_estimate_baseline: 0.05
top_cnt: 200
side_cnt: 25
max_solve_cnt: 500 # Max Point for solve; highly influence performace
min_dist: 20 # for vworks
freq: 10 # frequence (Hz) of publish tracking result. At least 10Hz for good estimation. If set 0, the frequence will be same as raw image
F_threshold: 1 # ransac threshold (pixel)
show_track: 0 # publish tracking image as topic
show_track_id: 0
flow_back: 1 # perform forward and backward optical flow to improve feature tracking accuracy
enable_perf_output: 0
#optimization parameters
max_solver_time: 0.15 # max solver itration time (ms), to guarantee real time
max_num_iterations: 10 # max solver itrations, to guarantee real time
keyframe_parallax: 12.0 # keyframe selection threshold (pixel)
#imu parameters The more accurate parameters you provide, the better performance
acc_n: 0.000945275948987 # accelerometer measurement noise standard deviation. #0.2 0.04
gyr_n: 0.00160297909721 # gyroscope measurement noise standard deviation. #0.05 0.004
acc_w: 2.16684207108e-05 # accelerometer bias random work noise standard deviation. #0.02
gyr_w: 3.23006936842e-05 # gyroscope bias random work noise standard deviation. #4.0e-5
g_norm: 9.81 # gravity magnitude
#unsynchronization parameters
estimate_td: 1 # online estimate time offset between camera and imu
td: -0.034769903289 #Use mynteye imu
#loop closure parameters
load_previous_pose_graph: 0 # load and reuse previous pose graph; load from 'pose_graph_save_path'
pose_graph_save_path: "/home/rui/output/pose_graph/" # save and load path
save_image: 0
any suggestions? many thanks!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.