You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
first of all, nice work, I really enjoyed using this stack 👍 I got it all running after a few system dependent hacks (got no PS3 controller, had to change the buttons). Anyway, I printed the Falkor logo, as provided, on a sheet of paper, inverted, which means: black logo on white paper. If I am right, this shouldn't harm the LBP classifier right? It's based on edge detection and the image will most probably converted to BW anyway...
As soon as I launch the drone (2.0), set it to "tracking" mode...it does nothing really, just hovering. The CV window also indicates "Tracking". I pulled the master, by the way, and I am using ROS Groovy on Ubuntu 12.10 (and also 12.04).
The only error messages are:
[ERROR] [1367779163.189744376]: Unable to open camera calibration file [/home/flier/.ros/camera_info/ardrone_front.yaml]
[ERROR] [1367779163.189816043]: Unable to open camera calibration file [/home/flier/.ros/camera_info/ardrone_bottom.yaml]
but they seem to be harmless, according to issue #2.
So - does the logo really need to be white on black? (I can see the optical flow bars though in the CV window...) Thank you for your time ;)
The text was updated successfully, but these errors were encountered:
Hi, not sure what you mean when you say you see the optical flow bars? Do you see a yellow box around the logo? If so it is detecting and the problem is in the control loop. If not then it isnt detecting. I have actually not had good luck with inverted images. If its not detecting try white on black.
first of all, nice work, I really enjoyed using this stack 👍 I got
it all running after a few system dependent hacks (got no PS3
controller, had to change the buttons). Anyway, I printed the Falkor
logo, as provided, on a sheet of paper, inverted, which means: black
logo on white paper. If I am right, this shouldn't harm the LBP
classifier right? It's based on edge detection and the image will most
probably converted to BW anyway...
As soon as I launch the drone (2.0), set it to "tracking" mode...it
does nothing really, just hovering. The CV window also indicates
"Tracking". I pulled the master, by the way, and I am using ROS Groovy
on Ubuntu 12.10 (and also 12.04).
The only error messages are:
[ERROR] [1367779163.189744376]: Unable to open camera calibration file
[/home/flier/.ros/camera_info/ardrone_front.yaml]
[ERROR] [1367779163.189816043]: Unable to open camera calibration file
[/home/flier/.ros/camera_info/ardrone_bottom.yaml]
but they seem to be harmless, according to issue #2.
So - does the logo really need to be white on black? (I can see the
optical flow bars though in the CV window...) Thank you for your time
;)
Reply to this email directly or view it on GitHub: #3
Hi there,
first of all, nice work, I really enjoyed using this stack 👍 I got it all running after a few system dependent hacks (got no PS3 controller, had to change the buttons). Anyway, I printed the Falkor logo, as provided, on a sheet of paper, inverted, which means: black logo on white paper. If I am right, this shouldn't harm the LBP classifier right? It's based on edge detection and the image will most probably converted to BW anyway...
As soon as I launch the drone (2.0), set it to "tracking" mode...it does nothing really, just hovering. The CV window also indicates "Tracking". I pulled the master, by the way, and I am using ROS Groovy on Ubuntu 12.10 (and also 12.04).
The only error messages are:
[ERROR] [1367779163.189744376]: Unable to open camera calibration file [/home/flier/.ros/camera_info/ardrone_front.yaml]
[ERROR] [1367779163.189816043]: Unable to open camera calibration file [/home/flier/.ros/camera_info/ardrone_bottom.yaml]
but they seem to be harmless, according to issue #2.
So - does the logo really need to be white on black? (I can see the optical flow bars though in the CV window...) Thank you for your time ;)
The text was updated successfully, but these errors were encountered: