Group Merging Very Slow. Also high RAM usage?

[BenjiDayan]

Hi @laughtervv ,

I've been trying to run your pretrained S3DIS model on point clouds from the matterport dataset. However I've found that group merging is very slow, for instance on a room divided into about 80 blocks, around 10000 points per block, group merging takes 5 mins + per block (if running on one cpu). Maybe I'm doing something wrong, did you also encounter this?

A separate problem is that to load one room into memory can take 30+ GB RAM, though I guess you could store and load on demand by block the model's output?

Sorry for all the questions, but also I've found group instance segmentation results to be a bit shoddy, perhaps it's something to do with the mingroupsize.txt and the pergroup_thres.txt and other thresholds? How did you get these values? I'm currently using the ones you posted on Issue #8.

Thanks for any help.

[laughtervv]

Group merging might be slow on large point cloud.
I preloaded the data into ram before training, so it may take time. You may load data in an online fashion and use cache (refer to here).
valid.py is used to generate mingroupsize.txt and the pergroup_thres.txt.