DensifyPointCloud Linux vs Windows (time ejecution)

[Livan89]

Hi.

Do you have any idea if the execution times of densifyPointCloud.exe on Linux are better or worse than on Windows? It refers to a compilation made in Linux and another in Windows

Thank you.

[cdcseacave]

some people reported slower execution on linux, but I think in the end it is only a matter of figuring out the right optimization flags to be set to the compiler; I am not a linux user, and I can not make any suggestion here

[rgherta]

Just ran DensifyPointCloud on ubuntu 19 on the dataset https://github.com/openMVG/ImageDataset_SceauxCastle

My logs are below, no nvidia or cuda

8. Densify point cloud

02:46:04 [App ] Build date: Jan 5 2020, 02:34:42 02:46:04 [App ] CPU: Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz (8 cores) 02:46:04 [App ] RAM: 7.67GB Physical Memory 976.00MB Virtual Memory 02:46:04 [App ] OS: Linux 5.3.0-26-generic (x86_64) 02:46:04 [App ] SSE & AVX compatible CPU & OS detected

02:46:04 [App ] Command line: scene.mvs -w /home/badwolf/Documents/webapps/opencv/BetaFace/out/mvs 02:46:04 [App ] Camera model loaded: platform 0; camera 0; f 1.054x1.054; poses 11 02:46:04 [App ] Image loaded 0: 100_7100.JPG 02:46:04 [App ] Image loaded 1: 100_7101.JPG 02:46:04 [App ] Image loaded 2: 100_7102.JPG 02:46:04 [App ] Image loaded 3: 100_7103.JPG 02:46:04 [App ] Image loaded 4: 100_7104.JPG 02:46:04 [App ] Image loaded 5: 100_7105.JPG 02:46:04 [App ] Image loaded 6: 100_7106.JPG 02:46:04 [App ] Image loaded 7: 100_7107.JPG 02:46:04 [App ] Image loaded 8: 100_7108.JPG 02:46:04 [App ] Image loaded 9: 100_7109.JPG 02:46:04 [App ] Image loaded 10: 100_7110.JPG 02:46:04 [App ] Scene loaded from interface format (13ms): 11 images (11 calibrated) with a total of 63.22 MPixels (5.75 MPixels/image) 6086 points, 0 vertices, 0 faces

02:46:04 [App ] Preparing images for dense reconstruction completed: 11 images (190ms) 02:46:04 [App ] Selecting images for dense reconstruction completed: 11 images (6ms)

Estimated depth-maps 3 (27.27%, 1m44s, ETA 4m)... Estimated depth-maps 7 (63.64%, 3m57s, ETA 2m)... Estimated depth-maps 11 (100%, 6m24s6ms)
Filtered depth-maps 11 (100%, 9s193ms)
Fused depth-maps 11 (100%, 2s951ms)

02:52:44 [App ] Depth-maps fused and filtered: 11 depth-maps, 7695674 depths, 1122541 points (15%%) (7s134ms) 02:52:45 [App ] Densifying point-cloud completed: 1122541 points (6m40s641ms) 02:52:48 [App ] Scene saved (3s76ms): 11 images (11 calibrated) 1122541 points, 0 vertices, 0 faces

02:52:48 [App ] Point-cloud saved: 1122541 points (165ms) 02:52:48 [App ] MEMORYINFO: { 02:52:48 [App ] VmPeak: 2909416 kB 02:52:48 [App ] VmSize: 2425412 kB 02:52:48 [App ] } ENDINFO