maekjuplease

Results 9 comments of maekjuplease

This issue can be fixed by copying the folder "camera_models" from original VINS-Fusion and paste to DynaVINS package. Refer and put in the same place as VINS-Fusion

Thank you for your response. I actually did as the same method that my semantic and RGB image topics have the same timestamps. However, when I launch the VIO, IMU...

Thank you so much for your response.

> Can you give example/screenshot of what you mean by "two cameras detect the lines"? There are some lines on the floor surface, I would like to segment those lines...

> For the masked images, what are they? Masked images are images only contains the lines

Here is the attachment. ![image](https://github.com/user-attachments/assets/23d7c189-24e2-4452-bd6d-d55d5d60614b)

https://github.com/introlab/rtabmap_ros/issues/871#top

Hi @matlabbe, I have tried this in the morning. However there is nothing output from pointcloud_to_depthimage node. After checking, this could be the problem from input format where my camera_info...

> To make [pointcloud_to_depthimage](http://wiki.ros.org/rtabmap_util#rtabmap_util.2Fpointcloud_to_depthimage) works correctly, it requires a valid `sensor_msgs/CameraInfo` using camera optical `frame_id`, with TF/URDF between camera and lidar accurate. In your case, the point cloud seems to...