Pro 2.1.3 some bruteforce estimator takes ages during Estimating Camera Location

[Mariusz_M]

Hi,

I put 200 images into Metashape Pro and tried to align them with preselection using Medium accuracy and default settings. On my computer it should take up to 5 minutes. But it stopped at 0% during Camera estimation and keeps working on some "bruteforce estimator:".... for ages. I actually tried to cancell the whole thing but it was cancelling while doing it for 25 minutes and was still counting, so I had to kill the process and restart Metashape.

First I thought it may have had something to do with a few masks I put on the images in the first attempt. So second time around I used the same 200 images and put one sma rectangular mask on one image and run alignment again. Same outcome... bruteforce estimator for ages...

Third time around I switched the photosed and used something similar, also 200 images, but different images and no masks at all... Still the same... Already running for over 25 minutes and still at 0% of camera estimation.

I have used these photosets before with previous Metashape Versions. So nothing wrong with pictures, but something clearly wrong with the software. Is there any tweak to switch off this "bruteforce estimator:" and keep using the current version like it was an old version?

This is what I see in the console:

pair 3 and 38: 784 robust from 4943
pair 1 and 41: 1931 robust from 4954
pair 4 and 9: 1025 robust from 4928
additional pairs considered in 3.596 sec.
optimizing initial pair...
****************************************************************************************************
default pair score: n_aligned: 3, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 0.935702, accuracy: 0.00583939, n_points: 5757, ids: [1, 2]
optimal pair score: n_aligned: 3, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 0.683379, accuracy: 0.00306118, n_points: 3648, ids: [36, 38]
initial pair still not stable
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.91284, accuracy: 0.00280702, n_points: 6811, ids: [36, 38]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.90761, accuracy: 0.00280855, n_points: 6802, ids: [36, 38]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.90067, accuracy: 0.00280848, n_points: 6803, ids: [36, 38]
bruteforce estimator: n skipped: 5 / 40, tm_total = 63.3, tm0 = 0.004, tm1 = 0, tm2 = 61.67
initial pair 1/20 : t_calculate: 63.315, t_evaluate: 3.107
new best bruteforce evaluated score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.89912, accuracy: 0.00280921, n_points: 6804, ids: [36, 38]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.93716, accuracy: 0.00279238, n_points: 6822, ids: [36, 37]
bruteforce estimator: n skipped: 12 / 34, tm_total = 14.777, tm0 = 0.004, tm1 = 0, tm2 = 13.103
initial pair 2/20 : t_calculate: 14.796, t_evaluate: 3.02
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.26831, accuracy: 0.00420025, n_points: 7414, ids: [4, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.25748, accuracy: 0.00420639, n_points: 7404, ids: [4, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.25487, accuracy: 0.00420779, n_points: 7403, ids: [4, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.25487, accuracy: 0.00420779, n_points: 7403, ids: [4, 37]
bruteforce estimator: n skipped: 4 / 88, tm_total = 172.921, tm0 = 0.009, tm1 = 0, tm2 = 169.768
initial pair 3/20 : t_calculate: 172.937, t_evaluate: 5.848
new best bruteforce evaluated score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.25487, accuracy: 0.00420779, n_points: 7403, ids: [4, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.23821, accuracy: 0.0114749, n_points: 8571, ids: [1, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.22273, accuracy: 0.0114948, n_points: 8570, ids: [1, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.20827, accuracy: 0.0115137, n_points: 8568, ids: [1, 2]
bruteforce estimator: n skipped: 6 / 38, tm_total = 79.193, tm0 = 0.007, tm1 = 0, tm2 = 76.577
initial pair 4/20 : t_calculate: 79.218, t_evaluate: 6.22
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.2351, accuracy: 0.0114733, n_points: 8568, ids: [0, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.22412, accuracy: 0.0115737, n_points: 8569, ids: [0, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.21902, accuracy: 0.0115979, n_points: 8569, ids: [0, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.2172, accuracy: 0.011533, n_points: 8568, ids: [0, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.20202, accuracy: 0.011615, n_points: 8565, ids: [0, 2]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.12348, accuracy: 0.0115042, n_points: 8534, ids: [0, 2]
bruteforce estimator: n skipped: 4 / 90, tm_total = 142.041, tm0 = 0.013, tm1 = 0, tm2 = 138.578
initial pair 5/20 : t_calculate: 142.055, t_evaluate: 5.147
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.77631, accuracy: 0.0023308, n_points: 7353, ids: [5, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.73912, accuracy: 0.00233812, n_points: 7340, ids: [5, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.27191, accuracy: 0.00419602, n_points: 7416, ids: [5, 37]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.26893, accuracy: 0.00420003, n_points: 7420, ids: [5, 37]
bruteforce estimator: n skipped: 1 / 58, tm_total = 113.435, tm0 = 0.008, tm1 = 0, tm2 = 111.154
initial pair 6/20 : t_calculate: 113.451, t_evaluate: 5.2
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.93273, accuracy: 0.00279949, n_points: 6823, ids: [6, 36]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.90957, accuracy: 0.00280206, n_points: 6814, ids: [6, 36]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.90535, accuracy: 0.00280512, n_points: 6807, ids: [6, 36]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.48033, accuracy: 0.00231413, n_points: 7099, ids: [6, 36]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.4642, accuracy: 0.00231315, n_points: 7094, ids: [6, 36]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.40826, accuracy: 0.00231871, n_points: 7064, ids: [6, 36]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.37616, accuracy: 0.00232043, n_points: 7046, ids: [6, 36]
bruteforce estimator: n skipped: 10 / 100, tm_total = 150.656, tm0 = 0.012, tm1 = 0, tm2 = 146.891
initial pair 7/20 : t_calculate: 150.674, t_evaluate: 4.134
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.24525, accuracy: 0.011531, n_points: 8574, ids: [0, 1]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 2.20229, accuracy: 0.0115619, n_points: 8566, ids: [0, 1]
bruteforce estimator: n skipped: 7 / 44, tm_total = 83.581, tm0 = 0.007, tm1 = 0, tm2 = 80.927
initial pair 8/20 : t_calculate: 83.602, t_evaluate: 5.109
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.741, accuracy: 0.00980469, n_points: 7595, ids: [71, 72]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.741, accuracy: 0.00980469, n_points: 7595, ids: [71, 72]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.741, accuracy: 0.00980469, n_points: 7595, ids: [71, 72]
bruteforce estimator: n skipped: 5 / 34, tm_total = 55.448, tm0 = 0.003, tm1 = 0, tm2 = 54.28
initial pair 9/20 : t_calculate: 55.46, t_evaluate: 5.101
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.71491, accuracy: 0.0170769, n_points: 7128, ids: [21, 22]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.65335, accuracy: 0.0172729, n_points: 7088, ids: [21, 22]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.64756, accuracy: 0.0172456, n_points: 7090, ids: [21, 22]
bruteforce estimator: n skipped: 10 / 96, tm_total = 113.377, tm0 = 0.011, tm1 = 0, tm2 = 109.557
initial pair 10/20 : t_calculate: 113.399, t_evaluate: 4.322
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.81521, accuracy: 0.0116906, n_points: 8408, ids: [2, 4]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.78542, accuracy: 0.011712, n_points: 8389, ids: [2, 4]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.76972, accuracy: 0.0117169, n_points: 8384, ids: [2, 4]
bruteforce estimator: new best score: n_aligned: 5, n_points_tier: 1, accuracy_tier: 1, reprojection_error: 1.76804, accuracy: 0.0117205, n_points: 8382, ids: [2, 4]

[Mariusz_M]

Any thoughts? I need to start working on these photosets.

[Alexey Pasumansky]

Hello Mariusz,

Can you share to support@agisoft.com the download link to the sample dataset that we could analyze on our side? Would be good to have set of original images and Metashape project with the alignment results (if it ever finishes) or, as an alternative to the project file, at least the screenshot of the Align Photos dialog with the parameters used.