mesh_extract error!Segmentation fault

[zhp2018]

hi,thank for your great work!
I train on my own data,but mesh_extract error:
(geosvr) zhp@DESKTOP-PVNH71K:~/code/GeoSVR$ python mesh_extract/tsdf_mesh.py ./output/ Rendering ./output/ Read dataset in COLMAP format. images_folder=./data/desktop/images res_downscale=0.0 res_width=0 [WARN] Source images are too large (2702x3597). Auto downscale gt by 3.0444786625197566. Use --images, --res_downscale, or --res_width to prevent it. computing nearest_id Rendering progress: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 28/28 [00:00<00:00, 142.40it/s] TSDF Fusion progress: 0%| | 0/28 [00:00<?, ?it/s] Segmentation fault (core dumped)

[Fictionarry]

Hi, this error is often raised when the RAM is exhausted. The TSDF fusion is costly for it. You may check if it is the problem. If so, increase --voxel_size or decrease --max_depth may help.

[zhp2018]

thanks!I'll try it