conf parameters explanation

[raghad1234]

Hi,
I just have two questions about the parameters used to reconstruct the images in your pkg
1- What do the following parameters represent and is there a way (other than trial and error) to set them? also what are the units of each?
--dimZ
--adaptive_threshold_c
--median_filter_size
--radius_search
--min_depth
--max_depth
2- Can the time parameters be set automatically?
--start_time_s=4.2
--stop_time_s=6.2

Thanks in advance

[guillermogb]

1. See the source code

2. The time depends on the scene and amount of motion. Typically, you want to process the last N events (for example, 1 million events). If you process more events and there is noise in the camera poses, it will show up in the disparity space image (DSI), and ultimately, in the reconstructed semi-dense depth map. If you process too few events, you will not have enough baseline or good signal-to-noise ratio, so the reconstructions will not be good.

[raghad1234]

Thanks for your reply. One more question, I am trying to extract the Eucludian clusters of the generated pcd by your pkg but I am getting a big failure and I think this is due to the low density of the pcd file. Do you recommend any method to cluster the different objects and classify them?
Actually, my end goal is to find the pose of the objects detected by DAVIS and the traditional ways of doing so are not working so far.
Really appreciate your input in this manner!