I hope this message finds you well. First, please allow me to apologize for reaching out after so many years since the project was published. However, I am deeply fascinated by your purely geometry-based method for depth map segmentation.
In my practical experiments, I encountered an issue when using a consumer-grade RGB-D camera to record ROS bags. I noticed that at slightly greater distances, the depth segmentation results become highly fragmented, leading to significantly degraded mapping performance. This problem only seems to disappear when the distance is within approximately 1 meter, at which point the depth segmentation becomes relatively stable.
I would be extremely grateful if you could offer any advice on how to resolve or mitigate this issue. Thank you so much in advance for your time and help!
Best regards.
I hope this message finds you well. First, please allow me to apologize for reaching out after so many years since the project was published. However, I am deeply fascinated by your purely geometry-based method for depth map segmentation.
In my practical experiments, I encountered an issue when using a consumer-grade RGB-D camera to record ROS bags. I noticed that at slightly greater distances, the depth segmentation results become highly fragmented, leading to significantly degraded mapping performance. This problem only seems to disappear when the distance is within approximately 1 meter, at which point the depth segmentation becomes relatively stable.
I would be extremely grateful if you could offer any advice on how to resolve or mitigate this issue. Thank you so much in advance for your time and help!
Best regards.