Revision history [back]

The stereo_view node does not allow for direct manipulation of the image size. The autosize parameter only determines whether the window will be fixed and sized to the image (autosize = true), or the user will be able to change the size (autosize = false). To change the parameter, you can write a launch file to launch the node and set the parameter:

<launch>
  <node name="stereo_view" pkg="image_view" type="stereo_view">
    <param name="autosize" value="true" />
    <remap from="image" to="image_rect_color" />
  </node>
</launch>

Since you mention editing the window size, there are at least three ways to resize the window:

1. The ROS-iest way to do it is to use the image_proc/resize nodelet to make the image the size you want it to be before displaying it. You can then use the resized image as input to the stereo_view node and set autosize = true to force the output window to remain that size.

2. Another option would be to modify the stereo_view node source code (or write your own node). There you can use OpenCV commands (e.g., cv2.resize) to change the image/window size.

3. A (perhaps messier) option would be to write a separate script to resize the window (with autosize = false). For this option, I would suggest searching somewhere like Stack Overflow (e.g., this answer mentions the Python win32gui package).

The stereo_view node does not allow for direct manipulation of the image size. The autosize parameter only determines whether the window will be fixed and sized to the image (autosize = true), or the user will be able to change the size (autosize = false). To change the parameter, you can write a launch file to launch the node and set the parameter:

<launch>
  <node name="stereo_view" pkg="image_view" type="stereo_view">
    <param name="autosize" value="true" value="false" />
    <remap from="image" to="image_rect_color" />
  </node>
</launch>

Since you mention editing the window size, there are at least three ways to resize the window:

1. The ROS-iest way to do it is to use the image_proc/resize nodelet to make the image the size you want it to be before displaying it. You can then use the resized image as input to the stereo_view node and set autosize = true to force the output window to remain that size.

2. Another option would be to modify the stereo_view node source code (or write your own node). There you can use OpenCV commands (e.g., cv2.resize) to change the image/window size.

3. A (perhaps messier) option would be to write a separate script to resize the window (with autosize = false). For this option, I would suggest searching somewhere like Stack Overflow (e.g., this answer mentions the Python win32gui package).