Ask Your Question
0

Viso2_ros without camera_info

asked 2014-11-02 05:16:30 -0500

dottant gravatar image

Hi all, I'm trying to compute visual odometry using a stream of images (left and rigth) from a server These images come without camera_info, but i have a calibration file

cam0: T_cam_imu: - [0.01486554298177506, 0.9995572490081733, -0.025774436697406162, 0.06522290953548025] - [-0.9998809296982875, 0.014967213324687367, 0.003756188357965682, -0.020706385492743767] - [0.004140296794218436, 0.025715529947969162, 0.999660727177951, -0.008054602460032234] - [0.0, 0.0, 0.0, 1.0] cam_overlaps: [1] camera_model: pinhole distortion_coeffs: [-0.28340811217029355, 0.07395907389290132, 0.00019359502856909603, 1.7618711454538528e-05] distortion_model: radtan intrinsics: [458.6548807207614, 457.2966964634893, 367.2158039615726, 248.37534060980727] resolution: [752, 480] rostopic: /cam0/image_raw cam1: T_cam_imu: - [0.012555267089114197, 0.9995987811512158, -0.025389800891792866, -0.04490198068250312] - [-0.9997550997230579, 0.013011905181515238, 0.017900583825283624, -0.02056977125889907] - [0.018223771455430235, 0.025158836311545446, 0.999517347077773, -0.008638135126028613] - [0.0, 0.0, 0.0, 1.0] T_cn_cnm1: - [0.9999972564778362, 0.002312067192393073, 0.00037600810233414045, -0.11007380812712879] - [-0.0023171357232427163, 0.9998980485068706, 0.014089835846681338, 0.0003991215470570765] - [-0.00034339312053936386, -0.014090668452710595, 0.9999006626379018, -0.0008537025033537094] - [0.0, 0.0, 0.0, 1.0] cam_overlaps: [0] camera_model: pinhole distortion_coeffs: [-0.283683654496499, 0.07451284309294819, -0.00010473894909809367, -3.555907002742513e-05] distortion_model: radtan intrinsics: [457.5874266035361, 456.13442556023665, 379.9994465203525, 255.2381853862733] resolution: [752, 480] rostopic: /cam1/image_raw

Well, i know that i need to rectify the images (before send everything to viso2) with stereo_image_proc, but if i don't have the camera_info stream i can't do it. I just have the image_raw pair, nothing else. Do you have any suggests??? thanka a lot

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted
0

answered 2014-11-10 01:52:03 -0500

Miquel Massot gravatar image

Hi dottant,

for these cases we use a python script to publish the image files into the ros ecosystem using a camera info YAML file with the calibration.

The script is available inside our package srv_tools, and it is this one.

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

1 follower

Stats

Asked: 2014-11-02 05:16:30 -0500

Seen: 156 times

Last updated: Nov 10 '14