-
Notifications
You must be signed in to change notification settings - Fork 7
Expand file tree
/
Copy pathfaq.html
More file actions
1607 lines (1317 loc) · 102 KB
/
Copy pathfaq.html
File metadata and controls
1607 lines (1317 loc) · 102 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html lang="en" data-content_root="./" data-theme="auto">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<meta property="og:title" content="Frequently Asked Questions" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://colmap.github.io/faq.html" />
<meta property="og:site_name" content="COLMAP" />
<meta property="og:description" content="Adjusting the options for different reconstruction scenarios and output quality: COLMAP provides many options that can be tuned for different reconstruction scenarios and to trade off accuracy and ..." />
<meta property="og:image" content="https://colmap.github.io/_static/og-image.png" />
<meta property="og:image:alt" content="COLMAP — Structure-from-Motion & Multi-View Stereo" />
<meta name="description" content="Adjusting the options for different reconstruction scenarios and output quality: COLMAP provides many options that can be tuned for different reconstruction scenarios and to trade off accuracy and ..." />
<meta name="twitter:card" content="summary_large_image" />
<title>Frequently Asked Questions — COLMAP</title>
<script data-cfasync="false">
document.documentElement.dataset.mode = localStorage.getItem("mode") || "auto";
document.documentElement.dataset.theme = localStorage.getItem("theme") || "auto";
</script>
<!--
this give us a css class that will be invisible only if js is disabled
-->
<noscript>
<style>
.pst-js-only { display: none !important; }
</style>
</noscript>
<!-- Loaded before other Sphinx assets -->
<link href="_static/styles/theme.css?digest=90905a2f556bf617f1a9" rel="stylesheet" />
<link href="_static/styles/pydata-sphinx-theme.css?digest=90905a2f556bf617f1a9" rel="stylesheet" />
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=03e43079" />
<link rel="stylesheet" type="text/css" href="_static/sphinx-design.min.css?v=95c83b7e" />
<link rel="stylesheet" type="text/css" href="_static/custom.css?v=c4c8f65a" />
<link rel="stylesheet" type="text/css" href="_static/landing.css?v=1fe38c8d" />
<!-- So that users can add custom icons -->
<script defer src="_static/scripts/fontawesome.js?digest=90905a2f556bf617f1a9"></script>
<!-- Pre-loaded scripts that we'll load fully later -->
<link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=90905a2f556bf617f1a9" />
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=90905a2f556bf617f1a9" />
<script src="_static/documentation_options.js?v=72e9b2fb"></script>
<script src="_static/doctools.js?v=fd6eb6e6"></script>
<script src="_static/sphinx_highlight.js?v=6ffebe34"></script>
<script src="_static/design-tabs.js?v=f930bc37"></script>
<script>DOCUMENTATION_OPTIONS.pagename = 'faq';</script>
<script>DOCUMENTATION_OPTIONS.search_as_you_type = false;</script>
<script src="_static/install_selector.js?v=94a4b0a5"></script>
<script src="_static/external_links.js?v=cd4e6f29"></script>
<link rel="canonical" href="https://colmap.github.io/faq.html" />
<link rel="icon" href="_static/favicon.svg"/>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="Changelog" href="changelog.html" />
<link rel="prev" title="Cost Functions" href="pycolmap/cost_functions.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
<meta name="docsearch:version" content="4.2.0.dev0 | 109ebdb (2026-07-21)" />
<script src="_static/searchtools.js"></script>
<script src="_static/language_data.js"></script>
<script src="searchindex.js"></script>
</head>
<body data-default-mode="auto">
<div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
<div id="pst-scroll-pixel-helper"></div>
<button type="button" class="btn rounded-pill" id="pst-back-to-top">
<i class="fa-solid fa-arrow-up"></i>Back to top</button>
<dialog id="pst-search-dialog">
<form class="bd-search d-flex align-items-center"
action="search.html"
method="get">
<i class="fa-solid fa-magnifying-glass"></i>
<input type="search"
class="form-control"
name="q"
placeholder="Search the docs ..."
aria-label="Search the docs ..."
autocomplete="off"
autocorrect="off"
autocapitalize="off"
spellcheck="false"/>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form>
</dialog>
<div class="pst-async-banner-revealer d-none">
<aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
</div>
<header id="pst-header" class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
<div class="bd-header__inner bd-page-width">
<button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
<span class="fa-solid fa-bars"></span>
</button>
<div class=" navbar-header-items__start">
<div class="navbar-item">
<a class="navbar-brand logo" href="index.html">
<img src="_static/colmap-logo.svg" class="logo__image only-light" alt=""/>
<img src="_static/colmap-logo-dark.svg" class="logo__image only-dark pst-js-only" alt=""/>
<p class="title logo__title">COLMAP</p>
</a></div>
</div>
<div class=" navbar-header-items">
<div class="me-auto navbar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item ">
<a class="nav-link nav-internal" href="install.html">
Installation
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="tutorial.html">
Tutorial
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="concepts.html">
Key Concepts
</a>
</li>
<li class="nav-item dropdown">
<button class="btn dropdown-toggle nav-item" type="button"
data-bs-toggle="dropdown" aria-expanded="false"
aria-controls="pst-nav-more-links">
More
</button>
<ul id="pst-nav-more-links" class="dropdown-menu">
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="features.html">
Feature Extraction and Matching
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="database.html">
Database Format
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="cameras.html">
Camera Models
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="rigs.html">
Rig Support
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="format.html">
Output Format
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="datasets.html">
Datasets
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="gui.html">
Graphical User Interface
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="cli.html">
Command-line Interface
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="pycolmap/index.html">
PyCOLMAP
</a>
</li>
<li class=" current active">
<a class="nav-link dropdown-item nav-internal" href="#">
Frequently Asked Questions
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="changelog.html">
Changelog
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="contribution.html">
Contribution
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="license.html">
License
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="bibliography.html">
Bibliography
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="legacy.html">
Legacy Documentations
</a>
</li>
</ul>
</li>
</ul>
</nav></div>
</div>
<div class="navbar-header-items__end">
<div class="navbar-item navbar-persistent--container">
<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
</div>
<div class="navbar-item">
<div class="theme-switch-container dropdown pst-js-only" data-bs-toggle="tooltip" data-bs-placement="bottom" title="Color mode">
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button dropdown-toggle" aria-label="Color mode" data-bs-toggle="dropdown">
<i class="theme-switch fa-solid fa-sun fa-lg fa-fw" data-mode="light" title="Light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg fa-fw" data-mode="dark" title="Dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg fa-fw" data-mode="auto" title="System Settings"></i>
</button>
<ul class="dropdown-menu dropdown-menu-end">
<li><button class="dropdown-item d-flex align-items-center theme-change-button" data-mode="auto"><i class="fa-solid fa-circle-half-stroke fa-lg fa-fw me-1"></i>System Settings</button></li>
<li><button class="dropdown-item d-flex align-items-center theme-change-button" data-mode="light"><i class="fa-solid fa-sun fa-lg fa-fw me-1"></i>Light</button></li>
<li><button class="dropdown-item d-flex align-items-center theme-change-button" data-mode="dark"><i class="fa-solid fa-moon fa-lg fa-fw me-1"></i>Dark</button></li>
</ul>
</div></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/colmap/colmap" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-github fa-lg" aria-hidden="true"></i><span class="visually-hidden">GitHub</span></a>
</li>
<li class="nav-item">
<a href="https://pypi.org/project/pycolmap/" title="PyPI" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-python fa-lg" aria-hidden="true"></i><span class="visually-hidden">PyPI</span></a>
</li>
<li class="nav-item">
<a href="https://hub.docker.com/r/colmap/colmap" title="Docker Hub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-docker fa-lg" aria-hidden="true"></i><span class="visually-hidden">Docker Hub</span></a>
</li>
</ul></div>
</div>
</div>
<div class="navbar-persistent--mobile">
<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
</div>
<button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
<span class="fa-solid fa-outdent"></span>
</button>
</div>
</header>
<div class="bd-container">
<div class="bd-container__inner bd-page-width">
<dialog id="pst-primary-sidebar-modal"></dialog>
<div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
<div class="sidebar-header-items sidebar-primary__section">
<div class="sidebar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item ">
<a class="nav-link nav-internal" href="install.html">
Installation
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="tutorial.html">
Tutorial
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="concepts.html">
Key Concepts
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="features.html">
Feature Extraction and Matching
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="database.html">
Database Format
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="cameras.html">
Camera Models
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="rigs.html">
Rig Support
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="format.html">
Output Format
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="datasets.html">
Datasets
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="gui.html">
Graphical User Interface
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="cli.html">
Command-line Interface
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="pycolmap/index.html">
PyCOLMAP
</a>
</li>
<li class="nav-item current active">
<a class="nav-link nav-internal" href="#">
Frequently Asked Questions
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="changelog.html">
Changelog
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="contribution.html">
Contribution
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="license.html">
License
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="bibliography.html">
Bibliography
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="legacy.html">
Legacy Documentations
</a>
</li>
</ul>
</nav></div>
</div>
<div class="sidebar-header-items__end">
<div class="navbar-item">
<div class="theme-switch-container dropdown pst-js-only" data-bs-toggle="tooltip" data-bs-placement="bottom" title="Color mode">
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button dropdown-toggle" aria-label="Color mode" data-bs-toggle="dropdown">
<i class="theme-switch fa-solid fa-sun fa-lg fa-fw" data-mode="light" title="Light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg fa-fw" data-mode="dark" title="Dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg fa-fw" data-mode="auto" title="System Settings"></i>
</button>
<ul class="dropdown-menu dropdown-menu-end">
<li><button class="dropdown-item d-flex align-items-center theme-change-button" data-mode="auto"><i class="fa-solid fa-circle-half-stroke fa-lg fa-fw me-1"></i>System Settings</button></li>
<li><button class="dropdown-item d-flex align-items-center theme-change-button" data-mode="light"><i class="fa-solid fa-sun fa-lg fa-fw me-1"></i>Light</button></li>
<li><button class="dropdown-item d-flex align-items-center theme-change-button" data-mode="dark"><i class="fa-solid fa-moon fa-lg fa-fw me-1"></i>Dark</button></li>
</ul>
</div></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/colmap/colmap" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-github fa-lg" aria-hidden="true"></i><span class="visually-hidden">GitHub</span></a>
</li>
<li class="nav-item">
<a href="https://pypi.org/project/pycolmap/" title="PyPI" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-python fa-lg" aria-hidden="true"></i><span class="visually-hidden">PyPI</span></a>
</li>
<li class="nav-item">
<a href="https://hub.docker.com/r/colmap/colmap" title="Docker Hub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-docker fa-lg" aria-hidden="true"></i><span class="visually-hidden">Docker Hub</span></a>
</li>
</ul></div>
</div>
</div>
<div class="sidebar-primary-items__start sidebar-primary__section">
<div class="sidebar-primary-item pst-sidebar-collapse"><button id="pst-collapse-sidebar-button" aria-expanded="true" aria-controls="pst-primary-sidebar">
<svg class="pst-icon" role="img" aria-hidden="true" focusable="false" viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg">
<path fill="currentColor" d="M3 15.5C2.36232 15.5 1.74874 15.2564 1.28478 14.8189C0.820828 14.3815 0.541576 13.7832 0.504167 13.1467L0.5 13L0.5 3C0.499965 2.36232 0.743605 1.74874 1.18107 1.28478C1.61854 0.820828 2.21676 0.541576 2.85333 0.504167L3 0.5L13 0.5C13.6377 0.499965 14.2513 0.743605 14.7152 1.18107C15.1792 1.61854 15.4584 2.21676 15.4958 2.85333L15.5 3L15.5 13C15.5 13.6377 15.2564 14.2513 14.8189 14.7152C14.3815 15.1792 13.7832 15.4584 13.1467 15.4958L13 15.5L3 15.5ZM3 13.8333L10.5 13.8333L10.5 2.16667L3 2.16667C2.79589 2.16669 2.59889 2.24163 2.44636 2.37726C2.29383 2.5129 2.19638 2.69979 2.1725 2.9025L2.16667 3L2.16667 13C2.16669 13.2041 2.24163 13.4011 2.37726 13.5536C2.5129 13.7062 2.69979 13.8036 2.9025 13.8275L3 13.8333ZM6.65583 10.325L6.5775 10.2558L4.91083 8.58917C4.76735 8.44567 4.68116 8.25476 4.66843 8.05223C4.65569 7.84971 4.71729 7.6495 4.84167 7.48917L4.91083 7.41083L6.5775 5.74417C6.72747 5.59471 6.9287 5.50794 7.14032 5.50148C7.35194 5.49502 7.55809 5.56935 7.7169 5.70937C7.8757 5.8494 7.97525 6.04463 7.99533 6.25539C8.01541 6.46616 7.95451 6.67667 7.825 6.84417L7.75583 6.9225L6.67917 8L7.75583 9.0775C7.89931 9.22099 7.98551 9.41191 7.99824 9.61443C8.01097 9.81695 7.94938 10.0172 7.825 10.1775L7.75583 10.2558C7.61234 10.3993 7.42142 10.4855 7.2189 10.4982C7.01638 10.511 6.81617 10.4494 6.65583 10.325Z"/>
</svg>
<span class="pst-collapse-sidebar-label">Collapse Sidebar</span>
<span class="pst-expand-sidebar-label">Expand Sidebar</span>
</button></div>
<div class="sidebar-primary-item">
<nav class="bd-docs-nav bd-links"
aria-label="Section Navigation">
<p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
<div class="bd-toc-item navbar-nav"></div>
</nav></div>
</div>
<div class="sidebar-primary-items__end sidebar-primary__section">
<div class="sidebar-primary-item">
<div id="ethical-ad-placement"
class="flat"
data-ea-publisher="readthedocs"
data-ea-type="readthedocs-sidebar"
data-ea-manual="true">
</div></div>
</div>
</div>
<main id="main-content" class="bd-main" role="main">
<div class="bd-content">
<div class="bd-article-container">
<div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
<div class="header-article-items__start">
<div class="header-article-item">
<nav aria-label="Breadcrumb" class="d-print-none">
<ul class="bd-breadcrumbs">
<li class="breadcrumb-item breadcrumb-home">
<a href="index.html" class="nav-link" aria-label="Home">
<i class="fa-solid fa-home"></i>
</a>
</li>
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Frequently Asked Questions</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="frequently-asked-questions">
<h1>Frequently Asked Questions<a class="headerlink" href="#frequently-asked-questions" title="Link to this heading">#</a></h1>
<section id="adjusting-the-options-for-different-reconstruction-scenarios-and-output-quality">
<h2>Adjusting the options for different reconstruction scenarios and output quality<a class="headerlink" href="#adjusting-the-options-for-different-reconstruction-scenarios-and-output-quality" title="Link to this heading">#</a></h2>
<p>COLMAP provides many options that can be tuned for different reconstruction
scenarios and to trade off accuracy and completeness versus efficiency. The
default options are set for medium to high quality reconstruction of
unstructured input data. There are several presets for different scenarios and
quality levels, which can be set in the GUI as <code class="docutils literal notranslate"><span class="pre">Extras</span> <span class="pre">></span> <span class="pre">Set</span> <span class="pre">options</span> <span class="pre">for</span> <span class="pre">...</span></code>.
To use these presets from the command-line, you can save the current set of
options as <code class="docutils literal notranslate"><span class="pre">File</span> <span class="pre">></span> <span class="pre">Save</span> <span class="pre">project</span></code> after choosing the presets. The resulting
project file can be opened with a text editor to view the different options.
Alternatively, you can generate the project file also from the command-line
by running <code class="docutils literal notranslate"><span class="pre">colmap</span> <span class="pre">project_generator</span></code>.</p>
</section>
<section id="extending-colmap">
<h2>Extending COLMAP<a class="headerlink" href="#extending-colmap" title="Link to this heading">#</a></h2>
<p>If you need to simply analyze the produced sparse or dense reconstructions from
COLMAP, you can load the sparse models using pycolmap in Python or the
scripts in <code class="docutils literal notranslate"><span class="pre">scripts/matlab</span></code> for Matlab.</p>
<p>If you want to write a C/C++ executable that builds on top of COLMAP, there are
two possible approaches. First, the COLMAP headers and library are installed
to the <code class="docutils literal notranslate"><span class="pre">CMAKE_INSTALL_PREFIX</span></code> by default. Compiling against COLMAP as a
library is described <a class="reference internal" href="install.html#installation-library"><span class="std std-ref">here</span></a>. Alternatively, you can
start from the <code class="docutils literal notranslate"><span class="pre">src/colmap/tools/example.cc</span></code> code template and implement the desired
functionality directly as a new binary within COLMAP.</p>
</section>
<section id="choosing-between-sift-and-aliked-features">
<h2>Choosing between SIFT and ALIKED features<a class="headerlink" href="#choosing-between-sift-and-aliked-features" title="Link to this heading">#</a></h2>
<p>COLMAP supports two feature extraction algorithms: SIFT (default) and ALIKED
(requires ONNX support). Here are some guidelines for choosing between them:</p>
<ul class="simple">
<li><p><strong>SIFT</strong> is the most widely tested and robust choice. It works well for
scenarios with moderate to high view overlap, sufficient scene texture,
and captured under similar illumination conditions. It supports both GPU
and CPU extraction.</p></li>
<li><p><strong>ALIKED</strong> is a learned feature extractor that can produce more repeatable
features in some cases, particularly for scenes with limited view overlap,
little scene texture, and drastic illumination changes. It requires ONNX
Runtime at build time (<code class="docutils literal notranslate"><span class="pre">-DONNX_ENABLED=ON</span></code>).</p></li>
</ul>
<p>Both feature types support brute-force matching as well as LightGlue neural
network-based matching. LightGlue typically produces higher inlier ratios,
especially for image pairs with large viewpoint or illumination changes, but
requires ONNX support. See <a class="reference internal" href="features.html#features"><span class="std std-ref">Feature Extraction and Matching</span></a>
for details on available options.</p>
<p>Do not mix different feature types (e.g., SIFT and ALIKED) in the same
database, as the descriptors are incompatible.</p>
</section>
<section id="choosing-the-right-camera-model">
<span id="faq-choosing-camera-model"></span><h2>Choosing the right camera model<a class="headerlink" href="#choosing-the-right-camera-model" title="Link to this heading">#</a></h2>
<p>COLMAP supports many camera models with varying numbers of parameters (see
<a class="reference internal" href="cameras.html"><span class="doc">Camera Models</span></a> for the full list). Choosing the right model depends on your
lens type and reconstruction requirements:</p>
<ul class="simple">
<li><p><strong>SIMPLE_RADIAL</strong> (default): A good starting point for most standard cameras.
Models a single focal length, principal point, and one radial distortion
parameter.</p></li>
<li><p><strong>PINHOLE</strong>: Use if your images have negligible lens distortion (e.g.,
already undistorted images or high-quality industrial lenses).</p></li>
<li><p><strong>OPENCV</strong>: A good choice for wider-angle lenses with moderate distortion.
Models 2 focal lengths, principal point, and 4 distortion parameters (2 radial
+ 2 tangential).</p></li>
<li><p><strong>SIMPLE_RADIAL_FISHEYE</strong> or <strong>OPENCV_FISHEYE</strong>: Use for fisheye lenses with
a field of view significantly larger than 120 degrees.</p></li>
<li><p><strong>FULL_OPENCV</strong>: Use only when you have many images sharing intrinsics
and need to model complex distortion patterns. With 12 parameters, this model
requires a large number of observations to converge reliably.</p></li>
</ul>
<p>As a rule of thumb, use the simplest model that adequately describes your lens.
Overly complex models with many parameters can lead to degenerate or overfitted
calibration, especially when few images share intrinsics. If in doubt, start
with <code class="docutils literal notranslate"><span class="pre">SIMPLE_RADIAL</span></code> and inspect the reprojection errors in the model
statistics.</p>
</section>
<section id="using-calibration-from-opencv-kalibr-or-other-tools">
<h2>Using calibration from OpenCV, Kalibr, or other tools<a class="headerlink" href="#using-calibration-from-opencv-kalibr-or-other-tools" title="Link to this heading">#</a></h2>
<p>If you already calibrated your camera with an external tool such as OpenCV or
Kalibr, you can reuse those intrinsics in COLMAP (see <a class="reference internal" href="#faq-fix-intrinsics"><span class="std std-ref">Fix intrinsics</span></a> to keep them constant during reconstruction). Two
conventions have to be matched first.</p>
<p><strong>Pixel coordinate convention.</strong> COLMAP places the origin at the top-left
<em>corner</em> of the image, so the center of the top-left pixel is at <code class="docutils literal notranslate"><span class="pre">(0.5,</span> <span class="pre">0.5)</span></code>
and a centered principal point is <code class="docutils literal notranslate"><span class="pre">(width</span> <span class="pre">/</span> <span class="pre">2,</span> <span class="pre">height</span> <span class="pre">/</span> <span class="pre">2)</span></code>. OpenCV and Kalibr
place integer coordinates at pixel <em>centers</em>, so their centered principal point
is <code class="docutils literal notranslate"><span class="pre">((width</span> <span class="pre">-</span> <span class="pre">1)</span> <span class="pre">/</span> <span class="pre">2,</span> <span class="pre">(height</span> <span class="pre">-</span> <span class="pre">1)</span> <span class="pre">/</span> <span class="pre">2)</span></code>. To convert a principal point from
OpenCV/Kalibr to COLMAP, add <code class="docutils literal notranslate"><span class="pre">0.5</span></code> to both <code class="docutils literal notranslate"><span class="pre">cx</span></code> and <code class="docutils literal notranslate"><span class="pre">cy</span></code>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">cx_colmap</span> <span class="o">=</span> <span class="n">cx_opencv</span> <span class="o">+</span> <span class="mf">0.5</span>
<span class="n">cy_colmap</span> <span class="o">=</span> <span class="n">cy_opencv</span> <span class="o">+</span> <span class="mf">0.5</span>
</pre></div>
</div>
<p>For example, an OpenCV calibration of an 800×600 image with a centered principal
point <code class="docutils literal notranslate"><span class="pre">(399.5,</span> <span class="pre">299.5)</span></code> becomes <code class="docutils literal notranslate"><span class="pre">(400.0,</span> <span class="pre">300.0)</span></code> in COLMAP. The focal lengths
<code class="docutils literal notranslate"><span class="pre">fx</span></code>, <code class="docutils literal notranslate"><span class="pre">fy</span></code> and the distortion coefficients are unaffected by this shift.</p>
<p><strong>Distortion parameter order.</strong> COLMAP’s <code class="docutils literal notranslate"><span class="pre">OPENCV</span></code> model uses the same
<code class="docutils literal notranslate"><span class="pre">k1,</span> <span class="pre">k2,</span> <span class="pre">p1,</span> <span class="pre">p2</span></code> distortion parameters as OpenCV, and <code class="docutils literal notranslate"><span class="pre">FULL_OPENCV</span></code>
additionally uses <code class="docutils literal notranslate"><span class="pre">k3,</span> <span class="pre">k4,</span> <span class="pre">k5,</span> <span class="pre">k6</span></code>, in that order (see <a class="reference internal" href="cameras.html"><span class="doc">Camera Models</span></a>). A
camera line in <code class="docutils literal notranslate"><span class="pre">cameras.txt</span></code> for the example above is therefore:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="mi">1</span> <span class="n">OPENCV</span> <span class="mi">800</span> <span class="mi">600</span> <span class="n">fx</span> <span class="n">fy</span> <span class="mf">400.0</span> <span class="mf">300.0</span> <span class="n">k1</span> <span class="n">k2</span> <span class="n">p1</span> <span class="n">p2</span>
</pre></div>
</div>
</section>
<section id="choosing-between-incremental-global-and-hierarchical-sfm">
<h2>Choosing between incremental, global, and hierarchical SfM<a class="headerlink" href="#choosing-between-incremental-global-and-hierarchical-sfm" title="Link to this heading">#</a></h2>
<p>COLMAP offers three SfM pipelines:</p>
<ul class="simple">
<li><p><strong>Incremental mapper</strong> (<code class="docutils literal notranslate"><span class="pre">mapper</span></code>, default): Reconstructs the scene by
incrementally adding one image at a time. This is the most robust and
well-tested pipeline, but can become slow for large image collections, where
repeated bundle adjustment is often the bottleneck. This can be accelerated
substantially with the GPU-based Caspar backend (see
<a class="reference internal" href="#speedup-bundle-adjustment"><span class="std std-ref">Speedup bundle adjustment</span></a>).</p></li>
<li><p><strong>Global mapper</strong> (<code class="docutils literal notranslate"><span class="pre">global_mapper</span></code>): Solves for all camera poses
simultaneously using rotation averaging and global positioning. This can be
faster for large datasets with good matching graphs, but may be less robust
to outliers in the matching. The global mapper depends on good focal length
priors. If reliable intrinsics are not available, run
<code class="docutils literal notranslate"><span class="pre">view_graph_calibrator</span></code> before <code class="docutils literal notranslate"><span class="pre">global_mapper</span></code> to estimate them from the
view graph (optional but recommended to improve the quality of
global SfM). Note that <code class="docutils literal notranslate"><span class="pre">view_graph_calibrator</span></code> modifies the database
in-place, so it is recommended to work on a copy.</p></li>
<li><p><strong>Hierarchical mapper</strong> (<code class="docutils literal notranslate"><span class="pre">hierarchical_mapper</span></code>): Partitions the scene into
overlapping sub-models and reconstructs each independently, then merges them.
This is useful for very large-scale datasets where the incremental approach
becomes too slow but is usually less robust than the other two pipelines.</p></li>
</ul>
<p>All three can also be selected via the <code class="docutils literal notranslate"><span class="pre">automatic_reconstructor</span></code> using
<code class="docutils literal notranslate"><span class="pre">--mapper</span> <span class="pre">INCREMENTAL</span></code>, <code class="docutils literal notranslate"><span class="pre">--mapper</span> <span class="pre">GLOBAL</span></code>, or <code class="docutils literal notranslate"><span class="pre">--mapper</span> <span class="pre">HIERARCHICAL</span></code>.</p>
</section>
<section id="reconstruction-with-pose-priors-gps">
<h2>Reconstruction with pose priors (GPS)<a class="headerlink" href="#reconstruction-with-pose-priors-gps" title="Link to this heading">#</a></h2>
<p>If your images have GPS information in their EXIF metadata, COLMAP
automatically extracts and stores it as pose priors in the database during
feature extraction. These priors can then be used during reconstruction with
the <code class="docutils literal notranslate"><span class="pre">pose_prior_mapper</span></code>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap feature_extractor \
--database_path $PROJECT_PATH/database.db \
--image_path $PROJECT_PATH/images
colmap exhaustive_matcher \
--database_path $PROJECT_PATH/database.db
colmap pose_prior_mapper \
--database_path $PROJECT_PATH/database.db \
--image_path $PROJECT_PATH/images \
--output_path $PROJECT_PATH/sparse
</pre></div>
</div>
<p>The <code class="docutils literal notranslate"><span class="pre">pose_prior_mapper</span></code> is essentially the incremental mapper with prior
position constraints enabled. You can override the priors covariance (uncertainty)
using <code class="docutils literal notranslate"><span class="pre">--overwrite_priors_covariance</span></code>. The new covariance will be built based
on the values of <code class="docutils literal notranslate"><span class="pre">--prior_position_std_x</span></code>, <code class="docutils literal notranslate"><span class="pre">--prior_position_std_y</span></code>, and
<code class="docutils literal notranslate"><span class="pre">--prior_position_std_z</span></code> (default: 1.0 meter each).</p>
<p>For geo-registration of an already reconstructed model (without using priors
during mapping), see the <a class="reference internal" href="#geo-registration">Geo-registration</a> section.</p>
</section>
<section id="share-intrinsics">
<span id="faq-share-intrinsics"></span><h2>Share intrinsics<a class="headerlink" href="#share-intrinsics" title="Link to this heading">#</a></h2>
<p>COLMAP supports shared intrinsics for arbitrary groups of images and camera
models. Images share the same intrinsics, if they refer to the same camera, as
specified by the <code class="docutils literal notranslate"><span class="pre">camera_id</span></code> property in the database. You can add new cameras
and set shared intrinsics in the database management tool. Please, refer to
<a class="reference internal" href="tutorial.html#database-management"><span class="std std-ref">Database Management</span></a> for more information.</p>
</section>
<section id="set-known-camera-intrinsics">
<h2>Set known camera intrinsics<a class="headerlink" href="#set-known-camera-intrinsics" title="Link to this heading">#</a></h2>
<p>If the camera calibration is known a priori, the recommended way to provide it
is during feature extraction using the <code class="docutils literal notranslate"><span class="pre">ImageReader</span></code> options:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap feature_extractor \
--database_path $PROJECT_PATH/database.db \
--image_path $PROJECT_PATH/images \
--ImageReader.single_camera 1 \
--ImageReader.camera_model OPENCV \
--ImageReader.camera_params "fx,fy,cx,cy,k1,k2,p1,p2"
</pre></div>
</div>
<p>The parameters must be provided as a comma-separated list in the order defined
by the chosen camera model (see <a class="reference internal" href="cameras.html"><span class="doc">Camera Models</span></a>). In the GUI, the equivalent
settings can be found under <code class="docutils literal notranslate"><span class="pre">Processing</span> <span class="pre">></span> <span class="pre">Feature</span> <span class="pre">extraction</span> <span class="pre">></span> <span class="pre">Custom</span>
<span class="pre">parameters</span></code>. Use <code class="docutils literal notranslate"><span class="pre">--ImageReader.single_camera</span> <span class="pre">1</span></code> if all images were captured
by the same physical camera with identical settings, so that they share one
camera in the database (see <a class="reference internal" href="#share-intrinsics">Share intrinsics</a>).</p>
<p>To modify the intrinsics of an existing database, do not edit the SQLite tables
by hand (the parameters are stored as binary blobs of doubles), but use
pycolmap’s database API instead:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span><span class="w"> </span><span class="nn">pycolmap</span>
<span class="k">with</span> <span class="n">pycolmap</span><span class="o">.</span><span class="n">Database</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="s2">"path/to/database.db"</span><span class="p">)</span> <span class="k">as</span> <span class="n">db</span><span class="p">:</span>
<span class="n">camera</span> <span class="o">=</span> <span class="n">db</span><span class="o">.</span><span class="n">read_camera</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
<span class="n">camera</span><span class="o">.</span><span class="n">params</span> <span class="o">=</span> <span class="p">[</span><span class="n">fx</span><span class="p">,</span> <span class="n">fy</span><span class="p">,</span> <span class="n">cx</span><span class="p">,</span> <span class="n">cy</span><span class="p">,</span> <span class="n">k1</span><span class="p">,</span> <span class="n">k2</span><span class="p">,</span> <span class="n">p1</span><span class="p">,</span> <span class="n">p2</span><span class="p">]</span>
<span class="n">camera</span><span class="o">.</span><span class="n">has_prior_focal_length</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">db</span><span class="o">.</span><span class="n">update_camera</span><span class="p">(</span><span class="n">camera</span><span class="p">)</span>
</pre></div>
</div>
<p>Note that the provided parameters are still refined during bundle adjustment by
default. To keep them fixed during the reconstruction, see
<a class="reference internal" href="#faq-fix-intrinsics"><span class="std std-ref">Fix intrinsics</span></a>.</p>
</section>
<section id="fix-intrinsics">
<span id="faq-fix-intrinsics"></span><h2>Fix intrinsics<a class="headerlink" href="#fix-intrinsics" title="Link to this heading">#</a></h2>
<p>By default, COLMAP tries to refine the intrinsic camera parameters (except
principal point) automatically during the reconstruction. Usually, if there are
enough images in the dataset and you share the intrinsics between multiple
images, the estimated intrinsic camera parameters in SfM should be better than
parameters manually obtained with a calibration pattern.</p>
<p>However, sometimes COLMAP’s self-calibration routine might converge in
degenerate parameters, especially in case of the more complex camera models with
many distortion parameters. If you know the calibration parameters a priori, you
can fix different parameter groups during the reconstruction. Choose
<code class="docutils literal notranslate"><span class="pre">Reconstruction</span> <span class="pre">></span> <span class="pre">Reconstruction</span> <span class="pre">options</span> <span class="pre">></span> <span class="pre">Bundle</span> <span class="pre">Adj.</span> <span class="pre">></span> <span class="pre">refine_*</span></code> and check
which parameter group to refine or to keep constant. Even if you keep the
parameters constant during the reconstruction, you can refine the parameters in
a final global bundle adjustment by setting <code class="docutils literal notranslate"><span class="pre">Reconstruction</span> <span class="pre">></span> <span class="pre">Bundle</span> <span class="pre">adj.</span>
<span class="pre">options</span> <span class="pre">></span> <span class="pre">refine_*</span></code> and then running <code class="docutils literal notranslate"><span class="pre">Reconstruction</span> <span class="pre">></span> <span class="pre">Bundle</span> <span class="pre">adjustment</span></code>.</p>
</section>
<section id="principal-point-refinement">
<h2>Principal point refinement<a class="headerlink" href="#principal-point-refinement" title="Link to this heading">#</a></h2>
<p>By default, COLMAP keeps the principal point constant during the reconstruction,
as principal point estimation is an ill-posed problem in general. Once all
images are reconstructed, the problem is most often constrained enough that you
can try to refine the principal point in global bundle adjustment, especially
when sharing intrinsic parameters between multiple images. Please, refer to
<a class="reference internal" href="#faq-fix-intrinsics"><span class="std std-ref">Fix intrinsics</span></a> for more information.</p>
</section>
<section id="increase-number-of-matches-sparse-3d-points">
<h2>Increase number of matches / sparse 3D points<a class="headerlink" href="#increase-number-of-matches-sparse-3d-points" title="Link to this heading">#</a></h2>
<p>To increase the number of matches, you should use the more discriminative
DSP-SIFT features instead of plain SIFT and also estimate the affine feature
shape using the options: <code class="docutils literal notranslate"><span class="pre">--SiftExtraction.estimate_affine_shape=true</span></code> and
<code class="docutils literal notranslate"><span class="pre">--SiftExtraction.domain_size_pooling=true</span></code>. In addition, you should enable
guided feature matching using: <code class="docutils literal notranslate"><span class="pre">--FeatureMatching.guided_matching=true</span></code>.</p>
<p>By default, COLMAP ignores two-view feature tracks in triangulation, resulting
in fewer 3D points than possible. Triangulation of two-view tracks can in rare
cases improve the stability of sparse image collections by providing additional
constraints in bundle adjustment. To also triangulate two-view tracks, unselect
the option <code class="docutils literal notranslate"><span class="pre">Reconstruction</span> <span class="pre">></span> <span class="pre">Reconstruction</span> <span class="pre">options</span> <span class="pre">></span> <span class="pre">Triangulation</span> <span class="pre">></span>
<span class="pre">ignore_two_view_tracks</span></code>. If your images are taken from far distance with
respect to the scene, you can try to reduce the minimum triangulation angle.</p>
</section>
<section id="reconstruct-sparse-dense-model-from-known-camera-poses">
<h2>Reconstruct sparse/dense model from known camera poses<a class="headerlink" href="#reconstruct-sparse-dense-model-from-known-camera-poses" title="Link to this heading">#</a></h2>
<p>If the camera poses are known and you want to reconstruct a sparse or dense
model of the scene, you must first manually construct a sparse model by creating
a <code class="docutils literal notranslate"><span class="pre">cameras.txt</span></code>, <code class="docutils literal notranslate"><span class="pre">points3D.txt</span></code>, and <code class="docutils literal notranslate"><span class="pre">images.txt</span></code> under a new folder:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>+── path/to/manually/created/sparse/model
│ +── cameras.txt
│ +── images.txt
│ +── points3D.txt
</pre></div>
</div>
<p>The <code class="docutils literal notranslate"><span class="pre">points3D.txt</span></code> file should be empty while every other line in the <code class="docutils literal notranslate"><span class="pre">images.txt</span></code>
should also be empty, since the sparse features are computed, as described below. You can
refer to <a class="reference internal" href="format.html#output-format"><span class="std std-ref">this article</span></a> for more information about the structure of
a sparse model.</p>
<p>Example of images.txt:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="mi">1</span> <span class="mf">0.695104</span> <span class="mf">0.718385</span> <span class="o">-</span><span class="mf">0.024566</span> <span class="mf">0.012285</span> <span class="o">-</span><span class="mf">0.046895</span> <span class="mf">0.005253</span> <span class="o">-</span><span class="mf">0.199664</span> <span class="mi">1</span> <span class="n">image0001</span><span class="o">.</span><span class="n">png</span>
<span class="c1"># Make sure every other line is left empty</span>
<span class="mi">2</span> <span class="mf">0.696445</span> <span class="mf">0.717090</span> <span class="o">-</span><span class="mf">0.023185</span> <span class="mf">0.014441</span> <span class="o">-</span><span class="mf">0.041213</span> <span class="mf">0.001928</span> <span class="o">-</span><span class="mf">0.134851</span> <span class="mi">2</span> <span class="n">image0002</span><span class="o">.</span><span class="n">png</span>
<span class="mi">3</span> <span class="mf">0.697457</span> <span class="mf">0.715925</span> <span class="o">-</span><span class="mf">0.025383</span> <span class="mf">0.018967</span> <span class="o">-</span><span class="mf">0.054056</span> <span class="mf">0.008579</span> <span class="o">-</span><span class="mf">0.378221</span> <span class="mi">1</span> <span class="n">image0003</span><span class="o">.</span><span class="n">png</span>
<span class="mi">4</span> <span class="mf">0.698777</span> <span class="mf">0.714625</span> <span class="o">-</span><span class="mf">0.023996</span> <span class="mf">0.021129</span> <span class="o">-</span><span class="mf">0.048184</span> <span class="mf">0.004529</span> <span class="o">-</span><span class="mf">0.313427</span> <span class="mi">2</span> <span class="n">image0004</span><span class="o">.</span><span class="n">png</span>
</pre></div>
</div>
<p>Each image above must have the same <code class="docutils literal notranslate"><span class="pre">image_id</span></code> (first column) as in the database (next step).
This database can be inspected either in the GUI (under <code class="docutils literal notranslate"><span class="pre">Database</span> <span class="pre">management</span> <span class="pre">></span> <span class="pre">Processing</span></code>),
or, one can create a reconstruction with colmap and later export it as text in order to see
the images.txt file it creates.</p>
<p>To reconstruct a sparse map, you first have to recompute features from the
images of the known camera poses as follows:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap feature_extractor \
--database_path $PROJECT_PATH/database.db \
--image_path $PROJECT_PATH/images
</pre></div>
</div>
<p>If your known camera intrinsics have large distortion coefficients, you should
now manually copy the parameters from your <code class="docutils literal notranslate"><span class="pre">cameras.txt</span></code> to the database, such
that the matcher can leverage the intrinsics. Modifying the database is possible
in many ways, but an easy option is to use pycolmap’s database API.
Otherwise, you can skip this step and
simply continue as follows:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap exhaustive_matcher \ # or alternatively any other matcher
--database_path $PROJECT_PATH/database.db
colmap point_triangulator \
--database_path $PROJECT_PATH/database.db \
--image_path $PROJECT_PATH/images
--input_path path/to/manually/created/sparse/model \
--output_path path/to/triangulated/sparse/model
</pre></div>
</div>
<p>Note that the sparse reconstruction step is not necessary in order to compute
a dense model from known camera poses. Assuming you computed a sparse model
from the known camera poses, you can compute a dense model as follows:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap image_undistorter \
--image_path $PROJECT_PATH/images \
--input_path path/to/triangulated/sparse/model \
--output_path path/to/dense/workspace
colmap patch_match_stereo \
--workspace_path path/to/dense/workspace
colmap stereo_fusion \
--workspace_path path/to/dense/workspace \
--output_path path/to/dense/workspace/fused.ply
</pre></div>
</div>
<p>Alternatively, you can also produce a dense model without a sparse model as:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap image_undistorter \
--image_path $PROJECT_PATH/images \
--input_path path/to/manually/created/sparse/model \
--output_path path/to/dense/workspace
</pre></div>
</div>
<p>Since the sparse point cloud is used to automatically select neighboring images
during the dense stereo stage, you have to manually specify the source images,
as described <a class="reference internal" href="#faq-dense-manual-source"><span class="std std-ref">here</span></a>. The dense stereo stage
now also requires a manual specification of the depth range.</p>
<p>Finally, in this case, fusion will fail to successfully match points if min_num_pixels is
left at the default (greater than 1). So also set that parameter, as below:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>colmap patch_match_stereo \
--workspace_path path/to/dense/workspace \
--PatchMatchStereo.depth_min $MIN_DEPTH \
--PatchMatchStereo.depth_max $MAX_DEPTH
colmap stereo_fusion \
--workspace_path path/to/dense/workspace \
--StereoFusion.min_num_pixels 1 \
--output_path path/to/dense/workspace/fused.ply
</pre></div>
</div>
</section>