Failed to process dataset

Rob_Syvertsen · 28 October 2021 06:02

I’m just getting started with OpenDroneMap and am having troubles with the first test dataset I created using DroneDeploy on my Mavic 2 Pro.

If I pick about 8 photos, it will process, but the orthophoto is really just a smear and often incorrectly located on the map. About 50-100m away from where it was captured. I can sometimes get up to 15 photos into a dataset and it completes. I always come across a photo that triggers the dataset to fail. The smallest sample to fail is with the first 9 photos. The first 8 work, but a task that includes the ninth causes it to fail.

The photos are here. They have motion blur as I rushed to take advantage of a break in the rain this evening and did not set the exposure correctly, I let DroneDeploy manage it.
https://drive.google.com/drive/folders/1H4M-wPQ9zriIGzw5dKxkeZ8EWQmLFTJp?usp=sharing

DroneDeploy was able to process the full dataset and it’s blurry, but looks correctish. This was using their Enhanced3d mode on the default settings, 75% front overlap, 70% side overlap, 3m/s flight speed and mapping angle of -65 degrees. This was a crosshatch pattern with a perimeter.

I tried increasing the minimum number of features to 12000, turned on Verbose and Debug to see if there was any indication of why it failed in the output.

[INFO] Initializing ODM - Thu Oct 28 05:34:53 2021
[INFO] ==============
[INFO] auto_boundary: False
[INFO] boundary: {}
[INFO] build_overviews: False
[INFO] camera_lens: auto
[INFO] cameras: {}
[INFO] cog: True
[INFO] copy_to: None
[INFO] crop: 3
[INFO] debug: True
[INFO] dem_decimation: 1
[INFO] dem_euclidean_map: False
[INFO] dem_gapfill_steps: 3
[INFO] dem_resolution: 5
[INFO] depthmap_resolution: 640
[INFO] dsm: False
[INFO] dtm: False
[INFO] end_with: odm_postprocess
[INFO] fast_orthophoto: False
[INFO] feature_quality: high
[INFO] feature_type: sift
[INFO] force_gps: False
[INFO] gcp: None
[INFO] geo: None
[INFO] gps_accuracy: 10
[INFO] ignore_gsd: False
[INFO] matcher_distance: 0
[INFO] matcher_neighbors: 8
[INFO] matcher_type: flann
[INFO] max_concurrency: 12
[INFO] merge: all
[INFO] mesh_octree_depth: 11
[INFO] mesh_size: 200000
[INFO] min_num_features: 12000
[INFO] name: f58903c5-ae4f-41f3-89ec-abe3be1877fd
[INFO] optimize_disk_space: False
[INFO] orthophoto_compression: DEFLATE
[INFO] orthophoto_cutline: False
[INFO] orthophoto_kmz: False
[INFO] orthophoto_no_tiled: False
[INFO] orthophoto_png: False
[INFO] orthophoto_resolution: 5
[INFO] pc_classify: False
[INFO] pc_csv: False
[INFO] pc_ept: True
[INFO] pc_filter: 2.5
[INFO] pc_geometric: False
[INFO] pc_las: False
[INFO] pc_quality: medium
[INFO] pc_rectify: False
[INFO] pc_sample: 0
[INFO] pc_tile: False
[INFO] primary_band: auto
[INFO] project_path: /var/www/data
[INFO] radiometric_calibration: none
[INFO] rerun: None
[INFO] rerun_all: False
[INFO] rerun_from: None
[INFO] resize_to: 2048
[INFO] skip_3dmodel: False
[INFO] skip_band_alignment: False
[INFO] skip_report: False
[INFO] sm_cluster: None
[INFO] smrf_scalar: 1.25
[INFO] smrf_slope: 0.15
[INFO] smrf_threshold: 0.5
[INFO] smrf_window: 18.0
[INFO] split: 999999
[INFO] split_image_groups: None
[INFO] split_overlap: 150
[INFO] texturing_data_term: gmi
[INFO] texturing_keep_unseen_faces: False
[INFO] texturing_outlier_removal_type: gauss_clamping
[INFO] texturing_skip_global_seam_leveling: False
[INFO] texturing_skip_local_seam_leveling: False
[INFO] texturing_tone_mapping: none
[INFO] tiles: False
[INFO] time: False
[INFO] use_3dmesh: False
[INFO] use_exif: False
[INFO] use_fixed_camera_params: False
[INFO] use_hybrid_bundle_adjustment: False
[INFO] verbose: True
[INFO] ==============
[INFO] Running dataset stage
[INFO] Loading dataset from: /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/images
[DEBUG] /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/images
[INFO] Loading 9 images
[DEBUG] Loaded DJI_0465.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.22222744444445 | lon: -122.57027011111111 | alt: 97.0 | band: RGB (0)
[DEBUG] Loaded DJI_0460.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.222223944444444 | lon: -122.57088633333333 | alt: 96.0 | band: RGB (0)
[DEBUG] Loaded DJI_0464.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.222227000000004 | lon: -122.57039191666667 | alt: 97.0 | band: RGB (0)
[DEBUG] Loaded DJI_0467.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.2222285 | lon: -122.57001261111111 | alt: 96.0 | band: RGB (0)
[DEBUG] Loaded DJI_0463.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.22222686111111 | lon: -122.57051811111111 | alt: 97.0 | band: RGB (0)
[DEBUG] Loaded DJI_0461.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.222225333333334 | lon: -122.57076880555555 | alt: 96.0 | band: RGB (0)
[DEBUG] Loaded DJI_0468.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.22222941666667 | lon: -122.5698908611111 | alt: 97.0 | band: RGB (0)
[DEBUG] Loaded DJI_0462.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.222225305555554 | lon: -122.57063891666667 | alt: 97.0 | band: RGB (0)
[DEBUG] Loaded DJI_0466.JPG | camera: Hasselblad L1D-20c | dimensions: 2048 x 1365 | lat: 49.22222825 | lon: -122.57013863888889 | alt: 96.0 | band: RGB (0)
[INFO] Wrote images database: /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/images.json
[INFO] Found 9 usable images
[INFO] Parsing SRS header: WGS84 UTM 10N
[INFO] Finished dataset stage
[INFO] Running split stage
[INFO] Normal dataset, will process all at once.
[INFO] Finished split stage
[INFO] Running merge stage
[INFO] Normal dataset, nothing to merge.
[INFO] Finished merge stage
[INFO] Running opensfm stage
[INFO] Writing exif overrides
[INFO] Maximum photo dimensions: 2048px
[INFO] Photo dimensions for feature extraction: 1024px
[INFO] Altitude data detected, enabling it for GPS alignment
[INFO] [‘use_exif_size: no’, ‘flann_algorithm: KDTREE’, ‘feature_process_size: 1024’, ‘feature_min_frames: 12000’, ‘processes: 12’, ‘matching_gps_neighbors: 8’, ‘matching_gps_distance: 0’, ‘optimize_camera_parameters: yes’, ‘undistorted_image_format: tif’, ‘bundle_outlier_filtering_type: AUTO’, ‘sift_peak_threshold: 0.066’, ‘align_orientation_prior: vertical’, ‘triangulation_type: ROBUST’, ‘retriangulation_ratio: 2’, ‘feature_type: SIFT’, ‘use_altitude_tag: yes’, ‘align_method: auto’, ‘local_bundle_radius: 0’]
[INFO] Wrote reference_lla.json
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm extract_metadata “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
2021-10-28 05:34:54,233 INFO: Extracting EXIF for DJI_0460.JPG
2021-10-28 05:34:54,268 INFO: Extracting EXIF for DJI_0464.JPG
2021-10-28 05:34:54,304 INFO: Extracting EXIF for DJI_0467.JPG
2021-10-28 05:34:54,339 INFO: Extracting EXIF for DJI_0463.JPG
2021-10-28 05:34:54,374 INFO: Extracting EXIF for DJI_0461.JPG
2021-10-28 05:34:54,411 INFO: Extracting EXIF for DJI_0468.JPG
2021-10-28 05:34:54,446 INFO: Extracting EXIF for DJI_0462.JPG
2021-10-28 05:34:54,481 INFO: Extracting EXIF for DJI_0466.JPG
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm detect_features “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
2021-10-28 05:34:55,533 INFO: Capping memory usage to ~ 10477.470703125 MB
2021-10-28 05:34:55,533 INFO: Expecting to process 200 images.
2021-10-28 05:34:55,540 INFO: Reading data for image DJI_0465.JPG (queue-size=0
2021-10-28 05:34:55,540 INFO: Reading data for image DJI_0467.JPG (queue-size=0
2021-10-28 05:34:55,540 INFO: Reading data for image DJI_0468.JPG (queue-size=0
2021-10-28 05:34:55,614 INFO: Reading data for image DJI_0462.JPG (queue-size=1
2021-10-28 05:34:55,614 INFO: Extracting ROOT_SIFT features for image DJI_0468.JPG
2021-10-28 05:34:55,616 INFO: Reading data for image DJI_0460.JPG (queue-size=1
2021-10-28 05:34:55,617 INFO: Extracting ROOT_SIFT features for image DJI_0465.JPG
2021-10-28 05:34:55,617 INFO: Reading data for image DJI_0463.JPG (queue-size=1
2021-10-28 05:34:55,618 INFO: Extracting ROOT_SIFT features for image DJI_0467.JPG
2021-10-28 05:34:55,636 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,644 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,644 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,701 INFO: Reading data for image DJI_0466.JPG (queue-size=1
2021-10-28 05:34:55,702 INFO: Extracting ROOT_SIFT features for image DJI_0462.JPG
2021-10-28 05:34:55,702 INFO: Extracting ROOT_SIFT features for image DJI_0460.JPG
2021-10-28 05:34:55,702 INFO: Reading data for image DJI_0464.JPG (queue-size=1
2021-10-28 05:34:55,703 INFO: Reading data for image DJI_0461.JPG (queue-size=1
2021-10-28 05:34:55,704 INFO: Extracting ROOT_SIFT features for image DJI_0463.JPG
2021-10-28 05:34:55,749 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,755 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,756 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,813 INFO: Extracting ROOT_SIFT features for image DJI_0466.JPG
2021-10-28 05:34:55,821 INFO: Extracting ROOT_SIFT features for image DJI_0464.JPG
2021-10-28 05:34:55,826 INFO: Finished reading images
2021-10-28 05:34:55,827 INFO: Extracting ROOT_SIFT features for image DJI_0461.JPG
2021-10-28 05:34:55,839 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,850 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:55,870 DEBUG: Computing sift with threshold 0.066
2021-10-28 05:34:56,019 DEBUG: Found 440 points in 0.3821425437927246s
2021-10-28 05:34:56,019 DEBUG: reducing threshold
2021-10-28 05:34:56,020 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,073 DEBUG: Found 277 points in 0.4285554885864258s
2021-10-28 05:34:56,073 DEBUG: reducing threshold
2021-10-28 05:34:56,073 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,083 DEBUG: Found 414 points in 0.4389777183532715s
2021-10-28 05:34:56,083 DEBUG: reducing threshold
2021-10-28 05:34:56,083 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,142 DEBUG: Found 225 points in 0.3830862045288086s
2021-10-28 05:34:56,142 DEBUG: reducing threshold
2021-10-28 05:34:56,142 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,152 DEBUG: Found 250 points in 0.4026951789855957s
2021-10-28 05:34:56,153 DEBUG: reducing threshold
2021-10-28 05:34:56,153 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,167 DEBUG: Found 213 points in 0.4105870723724365s
2021-10-28 05:34:56,167 DEBUG: reducing threshold
2021-10-28 05:34:56,167 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,222 DEBUG: Found 327 points in 0.38350510597229004s
2021-10-28 05:34:56,223 DEBUG: reducing threshold
2021-10-28 05:34:56,223 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,232 DEBUG: Found 257 points in 0.3822946548461914s
2021-10-28 05:34:56,233 DEBUG: reducing threshold
2021-10-28 05:34:56,233 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,254 DEBUG: Found 272 points in 0.38361644744873047s
2021-10-28 05:34:56,254 DEBUG: reducing threshold
2021-10-28 05:34:56,255 DEBUG: Computing sift with threshold 0.044000000000000004
2021-10-28 05:34:56,319 DEBUG: Found 4663 points in 0.29894232749938965s
2021-10-28 05:34:56,319 DEBUG: reducing threshold
2021-10-28 05:34:56,319 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,355 DEBUG: Found 3663 points in 0.28139352798461914s
2021-10-28 05:34:56,355 DEBUG: reducing threshold
2021-10-28 05:34:56,355 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,387 DEBUG: Found 4606 points in 0.3040289878845215s
2021-10-28 05:34:56,388 DEBUG: reducing threshold
2021-10-28 05:34:56,388 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,436 DEBUG: Found 3083 points in 0.29358625411987305s
2021-10-28 05:34:56,436 DEBUG: reducing threshold
2021-10-28 05:34:56,436 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,443 DEBUG: Found 3130 points in 0.2903163433074951s
2021-10-28 05:34:56,445 DEBUG: reducing threshold
2021-10-28 05:34:56,445 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,457 DEBUG: Found 2787 points in 0.2891993522644043s
2021-10-28 05:34:56,457 DEBUG: reducing threshold
2021-10-28 05:34:56,457 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,511 DEBUG: Found 3277 points in 0.27831149101257324s
2021-10-28 05:34:56,511 DEBUG: reducing threshold
2021-10-28 05:34:56,512 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,517 DEBUG: Found 3949 points in 0.29387712478637695s
2021-10-28 05:34:56,517 DEBUG: reducing threshold
2021-10-28 05:34:56,517 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,520 DEBUG: Found 3230 points in 0.2652721405029297s
2021-10-28 05:34:56,520 DEBUG: reducing threshold
2021-10-28 05:34:56,520 DEBUG: Computing sift with threshold 0.029333333333333336
2021-10-28 05:34:56,684 DEBUG: Found 12105 points in 0.36455512046813965s
2021-10-28 05:34:56,684 DEBUG: done
2021-10-28 05:34:56,695 DEBUG: Found 10749 points in 0.3400537967681885s
2021-10-28 05:34:56,695 DEBUG: reducing threshold
2021-10-28 05:34:56,695 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,740 DEBUG: Found 11712 points in 0.35193896293640137s
2021-10-28 05:34:56,740 DEBUG: reducing threshold
2021-10-28 05:34:56,740 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,747 DEBUG: Found 10018 points in 0.31109023094177246s
2021-10-28 05:34:56,747 DEBUG: reducing threshold
2021-10-28 05:34:56,747 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,755 DEBUG: Found 10065 points in 0.3103823661804199s
2021-10-28 05:34:56,755 DEBUG: reducing threshold
2021-10-28 05:34:56,755 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,768 DEBUG: Found 9637 points in 0.31057143211364746s
2021-10-28 05:34:56,768 DEBUG: reducing threshold
2021-10-28 05:34:56,768 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,811 DEBUG: Found 10098 points in 0.2989368438720703s
2021-10-28 05:34:56,811 DEBUG: reducing threshold
2021-10-28 05:34:56,811 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,830 DEBUG: Found 9860 points in 0.3099088668823242s
2021-10-28 05:34:56,831 DEBUG: reducing threshold
2021-10-28 05:34:56,831 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:56,841 DEBUG: Found 10873 points in 0.3240809440612793s
2021-10-28 05:34:56,841 DEBUG: reducing threshold
2021-10-28 05:34:56,842 DEBUG: Computing sift with threshold 0.01955555555555556
2021-10-28 05:34:57,047 DEBUG: Found 14611 points in 0.35157227516174316s
2021-10-28 05:34:57,047 DEBUG: done
2021-10-28 05:34:57,103 DEBUG: Found 14005 points in 0.3559539318084717s
2021-10-28 05:34:57,103 DEBUG: done
2021-10-28 05:34:57,108 DEBUG: Found 15286 points in 0.36780738830566406s
2021-10-28 05:34:57,108 DEBUG: done
2021-10-28 05:34:57,110 DEBUG: Found 13942 points in 0.3544912338256836s
2021-10-28 05:34:57,110 DEBUG: done
2021-10-28 05:34:57,124 DEBUG: Found 14195 points in 0.3556554317474365s
2021-10-28 05:34:57,124 DEBUG: done
2021-10-28 05:34:57,159 DEBUG: Found 13756 points in 0.3281741142272949s
2021-10-28 05:34:57,159 DEBUG: done
2021-10-28 05:34:57,166 DEBUG: Found 14175 points in 0.35485029220581055s
2021-10-28 05:34:57,166 DEBUG: done
2021-10-28 05:34:57,194 DEBUG: Found 14690 points in 0.3527860641479492s
2021-10-28 05:34:57,195 DEBUG: done
2021-10-28 05:34:57,399 DEBUG: No segmentation for DJI_0468.JPG, no features masked.
2021-10-28 05:34:57,867 DEBUG: No segmentation for DJI_0465.JPG, no features masked.
2021-10-28 05:34:57,924 DEBUG: No segmentation for DJI_0460.JPG, no features masked.
2021-10-28 05:34:58,038 DEBUG: No segmentation for DJI_0463.JPG, no features masked.
2021-10-28 05:34:58,039 DEBUG: No segmentation for DJI_0464.JPG, no features masked.
2021-10-28 05:34:58,040 DEBUG: No segmentation for DJI_0466.JPG, no features masked.
2021-10-28 05:34:58,041 DEBUG: No segmentation for DJI_0461.JPG, no features masked.
2021-10-28 05:34:58,042 DEBUG: No segmentation for DJI_0462.JPG, no features masked.
2021-10-28 05:34:58,080 DEBUG: No segmentation for DJI_0467.JPG, no features masked.
Cont’d in next reply.

Rob_Syvertsen · 28 October 2021 06:02

Rest of the log output
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm match_features “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
2021-10-28 05:34:59,477 INFO: Matching 36 image pairs
2021-10-28 05:34:59,478 INFO: Computing pair matching with 12 processes
2021-10-28 05:34:59,755 DEBUG: No segmentation for DJI_0464.JPG, no features masked.
2021-10-28 05:34:59,762 DEBUG: No segmentation for DJI_0461.JPG, no features masked.
2021-10-28 05:34:59,767 DEBUG: No segmentation for DJI_0462.JPG, no features masked.
2021-10-28 05:34:59,771 DEBUG: No segmentation for DJI_0460.JPG, no features masked.
2021-10-28 05:34:59,792 DEBUG: No segmentation for DJI_0465.JPG, no features masked.
2021-10-28 05:35:00,075 DEBUG: No segmentation for DJI_0468.JPG, no features masked.
2021-10-28 05:35:00,089 DEBUG: No segmentation for DJI_0466.JPG, no features masked.
2021-10-28 05:35:00,098 DEBUG: No segmentation for DJI_0463.JPG, no features masked.
2021-10-28 05:35:00,114 DEBUG: No segmentation for DJI_0467.JPG, no features masked.
2021-10-28 05:35:00,657 DEBUG: Matching DJI_0461.JPG and DJI_0464.JPG. Matcher: FLANN (symmetric) T-desc: 1.164 T-robust: 0.005 T-total: 1.170 Matches: 512 Robust: 488 Success: True
2021-10-28 05:35:00,672 DEBUG: Matching DJI_0462.JPG and DJI_0464.JPG. Matcher: FLANN (symmetric) T-desc: 1.168 T-robust: 0.013 T-total: 1.188 Matches: 1790 Robust: 1767 Success: True
2021-10-28 05:35:00,710 DEBUG: Matching DJI_0464.JPG and DJI_0460.JPG. Matcher: FLANN (symmetric) T-desc: 1.216 T-robust: 0.009 T-total: 1.225 Matches: 98 Robust: 70 Success: True
2021-10-28 05:35:00,731 DEBUG: Matching DJI_0461.JPG and DJI_0468.JPG. Matcher: FLANN (symmetric) T-desc: 1.217 T-robust: 0.030 T-total: 1.247 Matches: 36 Robust: 11 Success: False
2021-10-28 05:35:00,758 DEBUG: Matching DJI_0461.JPG and DJI_0462.JPG. Matcher: FLANN (symmetric) T-desc: 1.224 T-robust: 0.021 T-total: 1.273 Matches: 4599 Robust: 4591 Success: True
2021-10-28 05:35:00,785 DEBUG: Matching DJI_0461.JPG and DJI_0465.JPG. Matcher: FLANN (symmetric) T-desc: 1.295 T-robust: 0.001 T-total: 1.296 Matches: 123 Robust: 108 Success: True
2021-10-28 05:35:00,803 DEBUG: Matching DJI_0461.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 1.287 T-robust: 0.031 T-total: 1.318 Matches: 31 Robust: 9 Success: False
2021-10-28 05:35:00,849 DEBUG: Matching DJI_0460.JPG and DJI_0463.JPG. Matcher: FLANN (symmetric) T-desc: 1.358 T-robust: 0.002 T-total: 1.361 Matches: 528 Robust: 505 Success: True
2021-10-28 05:35:00,879 DEBUG: Matching DJI_0464.JPG and DJI_0463.JPG. Matcher: FLANN (symmetric) T-desc: 1.358 T-robust: 0.026 T-total: 1.395 Matches: 4564 Robust: 4554 Success: True
2021-10-28 05:35:00,923 DEBUG: Matching DJI_0460.JPG and DJI_0465.JPG. Matcher: FLANN (symmetric) T-desc: 1.414 T-robust: 0.024 T-total: 1.438 Matches: 35 Robust: 10 Success: False
2021-10-28 05:35:01,053 DEBUG: Matching DJI_0462.JPG and DJI_0465.JPG. Matcher: FLANN (symmetric) T-desc: 1.560 T-robust: 0.005 T-total: 1.566 Matches: 553 Robust: 529 Success: True
2021-10-28 05:35:01,224 DEBUG: Matching DJI_0465.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 1.718 T-robust: 0.011 T-total: 1.737 Matches: 2084 Robust: 2060 Success: True
2021-10-28 05:35:01,237 DEBUG: Matching DJI_0464.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 0.568 T-robust: 0.002 T-total: 0.572 Matches: 635 Robust: 602 Success: True
2021-10-28 05:35:01,280 DEBUG: Matching DJI_0468.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.528 T-robust: 0.010 T-total: 0.543 Matches: 2107 Robust: 2091 Success: True
2021-10-28 05:35:01,343 DEBUG: Matching DJI_0468.JPG and DJI_0462.JPG. Matcher: FLANN (symmetric) T-desc: 0.472 T-robust: 0.021 T-total: 0.493 Matches: 33 Robust: 11 Success: False
2021-10-28 05:35:01,410 DEBUG: Matching DJI_0462.JPG and DJI_0460.JPG. Matcher: FLANN (symmetric) T-desc: 0.658 T-robust: 0.024 T-total: 0.687 Matches: 1802 Robust: 1783 Success: True
2021-10-28 05:35:01,421 DEBUG: Matching DJI_0461.JPG and DJI_0460.JPG. Matcher: FLANN (symmetric) T-desc: 0.584 T-robust: 0.038 T-total: 0.662 Matches: 4735 Robust: 4720 Success: True
2021-10-28 05:35:01,462 DEBUG: Matching DJI_0463.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 0.636 T-robust: 0.039 T-total: 0.676 Matches: 118 Robust: 97 Success: True
2021-10-28 05:35:01,468 DEBUG: Matching DJI_0461.JPG and DJI_0463.JPG. Matcher: FLANN (symmetric) T-desc: 0.618 T-robust: 0.040 T-total: 0.665 Matches: 1938 Robust: 1920 Success: True
2021-10-28 05:35:01,507 DEBUG: Matching DJI_0462.JPG and DJI_0463.JPG. Matcher: FLANN (symmetric) T-desc: 0.783 T-robust: 0.031 T-total: 0.835 Matches: 4619 Robust: 4604 Success: True
2021-10-28 05:35:01,511 DEBUG: Matching DJI_0468.JPG and DJI_0464.JPG. Matcher: FLANN (symmetric) T-desc: 0.630 T-robust: 0.001 T-total: 0.631 Matches: 144 Robust: 120 Success: True
2021-10-28 05:35:01,653 DEBUG: Matching DJI_0460.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 0.695 T-robust: 0.034 T-total: 0.729 Matches: 31 Robust: 10 Success: False
2021-10-28 05:35:01,826 DEBUG: Matching DJI_0468.JPG and DJI_0465.JPG. Matcher: FLANN (symmetric) T-desc: 0.584 T-robust: 0.002 T-total: 0.588 Matches: 663 Robust: 638 Success: True
2021-10-28 05:35:01,904 DEBUG: Matching DJI_0462.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 0.798 T-robust: 0.051 T-total: 0.850 Matches: 35 Robust: 11 Success: False
2021-10-28 05:35:01,954 DEBUG: Matching DJI_0468.JPG and DJI_0463.JPG. Matcher: FLANN (symmetric) T-desc: 0.491 T-robust: 0.039 T-total: 0.531 Matches: 38 Robust: 10 Success: False
2021-10-28 05:35:01,998 DEBUG: Matching DJI_0465.JPG and DJI_0463.JPG. Matcher: FLANN (symmetric) T-desc: 0.558 T-robust: 0.025 T-total: 0.588 Matches: 1860 Robust: 1833 Success: True
2021-10-28 05:35:02,004 DEBUG: Matching DJI_0461.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 0.628 T-robust: 0.032 T-total: 0.661 Matches: 28 Robust: 9 Success: False
2021-10-28 05:35:02,013 DEBUG: Matching DJI_0468.JPG and DJI_0460.JPG. Matcher: FLANN (symmetric) T-desc: 0.504 T-robust: 0.035 T-total: 0.540 Matches: 36 Robust: 10 Success: False
2021-10-28 05:35:02,014 DEBUG: Matching DJI_0465.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.671 T-robust: 0.039 T-total: 0.731 Matches: 4650 Robust: 4636 Success: True
2021-10-28 05:35:02,120 DEBUG: Matching DJI_0467.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.861 T-robust: 0.017 T-total: 0.895 Matches: 5165 Robust: 5155 Success: True
2021-10-28 05:35:02,121 DEBUG: Matching DJI_0460.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.575 T-robust: 0.034 T-total: 0.609 Matches: 25 Robust: 10 Success: False
2021-10-28 05:35:02,177 DEBUG: Matching DJI_0464.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.707 T-robust: 0.004 T-total: 0.714 Matches: 1942 Robust: 1913 Success: True
2021-10-28 05:35:02,182 DEBUG: Matching DJI_0463.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.672 T-robust: 0.001 T-total: 0.674 Matches: 528 Robust: 497 Success: True
2021-10-28 05:35:02,227 DEBUG: Matching DJI_0468.JPG and DJI_0467.JPG. Matcher: FLANN (symmetric) T-desc: 0.556 T-robust: 0.009 T-total: 0.574 Matches: 4827 Robust: 4818 Success: True
2021-10-28 05:35:02,343 DEBUG: Matching DJI_0464.JPG and DJI_0465.JPG. Matcher: FLANN (symmetric) T-desc: 0.494 T-robust: 0.013 T-total: 0.517 Matches: 4948 Robust: 4933 Success: True
2021-10-28 05:35:02,347 DEBUG: Matching DJI_0462.JPG and DJI_0466.JPG. Matcher: FLANN (symmetric) T-desc: 0.441 T-robust: 0.002 T-total: 0.442 Matches: 129 Robust: 94 Success: True
2021-10-28 05:35:02,349 INFO: Matched 36 pairs (brown-brown: 36) in 2.871481299982406 seconds (0.07976338055515145 seconds/pair).
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm create_tracks “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
2021-10-28 05:35:04,135 INFO: reading features
2021-10-28 05:35:04,563 DEBUG: Merging features onto tracks
2021-10-28 05:35:05,109 DEBUG: Good tracks: 25980
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm reconstruct “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
2021-10-28 05:35:07,159 INFO: Starting incremental reconstruction
2021-10-28 05:35:07,544 INFO: Starting reconstruction with DJI_0466.JPG and DJI_0467.JPG
2021-10-28 05:35:07,692 INFO: Two-view reconstruction inliers: 5293 / 5294
2021-10-28 05:35:08,114 INFO: Triangulated: 4395
2021-10-28 05:35:10,778 DEBUG: Ceres Solver Report: Iterations: 101, Initial cost: 2.143361e+01, Final cost: 2.053542e+01, Termination: NO_CONVERGENCE
2021-10-28 05:35:10,830 INFO: Removed outliers: 127
2021-10-28 05:35:10,885 INFO: -------------------------------------------------------
2021-10-28 05:35:10,917 INFO: DJI_0468.JPG resection inliers: 1410 / 2159
2021-10-28 05:35:10,956 INFO: Adding DJI_0468.JPG to the reconstruction
2021-10-28 05:35:11,158 INFO: Shots and/or GCPs are well-conditioned. Using naive 3D-3D alignment.
2021-10-28 05:35:13,928 DEBUG: Ceres Solver Report: Iterations: 101, Initial cost: 8.985870e+02, Final cost: 7.926401e+01, Termination: NO_CONVERGENCE
2021-10-28 05:35:14,031 INFO: Removed outliers: 429
2021-10-28 05:35:14,039 INFO: -------------------------------------------------------
2021-10-28 05:35:14,072 INFO: DJI_0465.JPG resection inliers: 1400 / 1894
2021-10-28 05:35:14,124 INFO: Adding DJI_0465.JPG to the reconstruction
2021-10-28 05:35:14,393 INFO: Re-triangulating
2021-10-28 05:35:14,393 INFO: Shots and/or GCPs are well-conditioned. Using naive 3D-3D alignment.
2021-10-28 05:35:16,174 DEBUG: Ceres Solver Report: Iterations: 49, Initial cost: 1.028862e+03, Final cost: 1.999603e+02, Termination: CONVERGENCE
2021-10-28 05:35:19,160 DEBUG: Ceres Solver Report: Iterations: 53, Initial cost: 7.729988e+02, Final cost: 3.302017e+02, Termination: CONVERGENCE
2021-10-28 05:35:19,273 INFO: Removed outliers: 560
2021-10-28 05:35:19,280 INFO: -------------------------------------------------------
2021-10-28 05:35:19,307 INFO: DJI_0464.JPG resection inliers: 1141 / 1212
2021-10-28 05:35:19,359 INFO: Adding DJI_0464.JPG to the reconstruction
2021-10-28 05:35:19,642 INFO: Shots and/or GCPs are well-conditioned. Using naive 3D-3D alignment.
2021-10-28 05:35:20,213 DEBUG: Ceres Solver Report: Iterations: 11, Initial cost: 6.255741e+02, Final cost: 3.393927e+02, Termination: CONVERGENCE
2021-10-28 05:35:20,342 INFO: Removed outliers: 332
2021-10-28 05:35:20,350 INFO: -------------------------------------------------------
2021-10-28 05:35:20,372 INFO: DJI_0463.JPG resection inliers: 718 / 806
2021-10-28 05:35:20,426 INFO: Adding DJI_0463.JPG to the reconstruction
2021-10-28 05:35:20,730 INFO: -------------------------------------------------------
2021-10-28 05:35:20,751 INFO: DJI_0462.JPG resection inliers: 627 / 697
2021-10-28 05:35:20,798 INFO: Adding DJI_0462.JPG to the reconstruction
2021-10-28 05:35:21,109 INFO: Shots and/or GCPs are well-conditioned. Using naive 3D-3D alignment.
2021-10-28 05:35:22,162 DEBUG: Ceres Solver Report: Iterations: 16, Initial cost: 1.078622e+03, Final cost: 4.636682e+02, Termination: CONVERGENCE
2021-10-28 05:35:22,350 INFO: Removed outliers: 234
2021-10-28 05:35:22,359 INFO: -------------------------------------------------------
2021-10-28 05:35:22,381 INFO: DJI_0461.JPG resection inliers: 672 / 719
2021-10-28 05:35:22,423 INFO: Adding DJI_0461.JPG to the reconstruction
2021-10-28 05:35:22,736 INFO: -------------------------------------------------------
2021-10-28 05:35:22,754 INFO: DJI_0460.JPG resection inliers: 725 / 766
2021-10-28 05:35:22,796 INFO: Adding DJI_0460.JPG to the reconstruction
2021-10-28 05:35:23,047 INFO: Re-triangulating
2021-10-28 05:35:23,048 INFO: Shots and/or GCPs are well-conditioned. Using naive 3D-3D alignment.
2021-10-28 05:35:28,532 DEBUG: Ceres Solver Report: Iterations: 71, Initial cost: 1.898867e+03, Final cost: 5.811573e+02, Termination: CONVERGENCE
2021-10-28 05:35:34,264 DEBUG: Ceres Solver Report: Iterations: 44, Initial cost: 1.345769e+03, Final cost: 6.876113e+02, Termination: CONVERGENCE
2021-10-28 05:35:34,476 INFO: Removed outliers: 350
2021-10-28 05:35:34,485 INFO: -------------------------------------------------------
2021-10-28 05:35:34,486 INFO: Shots and/or GCPs are well-conditioned. Using naive 3D-3D alignment.
2021-10-28 05:35:35,370 DEBUG: Ceres Solver Report: Iterations: 9, Initial cost: 6.531644e+02, Final cost: 6.271796e+02, Termination: CONVERGENCE
2021-10-28 05:35:35,581 INFO: Removed outliers: 170
2021-10-28 05:35:35,724 INFO: Reconstruction 0: 9 images, 13107 points
2021-10-28 05:35:35,724 INFO: 1 partial reconstructions in total.
[INFO] Export reconstruction stats
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm compute_statistics --diagram_max_points 100000 “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
2021-10-28 05:35:41,711 DEBUG: locator: <matplotlib.colorbar._ColorbarAutoLocator object at 0x7f86521c3a00>
2021-10-28 05:35:41,711 DEBUG: Using auto colorbar locator <matplotlib.colorbar._ColorbarAutoLocator object at 0x7f86521c3a00> on colorbar
2021-10-28 05:35:41,712 DEBUG: Setting pcolormesh
2021-10-28 05:35:42,545 DEBUG: locator: <matplotlib.colorbar._ColorbarAutoLocator object at 0x7f8651d82f10>
2021-10-28 05:35:42,545 DEBUG: Using auto colorbar locator <matplotlib.colorbar._ColorbarAutoLocator object at 0x7f8651d82f10> on colorbar
2021-10-28 05:35:42,545 DEBUG: Setting pcolormesh
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm export_geocoords --reconstruction --proj “+proj=utm +zone=10 +datum=WGS84 +units=m +no_defs +type=crs” --offset-x 531282.0 --offset-y 5452248.0 “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
[INFO] Updating /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/config.yaml
[INFO] undistorted_image_max_size: 1163
[INFO] Undistorting /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm …
2021-10-28 05:35:49,058 DEBUG: Undistorting image DJI_0460.JPG
2021-10-28 05:35:49,058 DEBUG: Undistorting image DJI_0461.JPG
2021-10-28 05:35:49,059 DEBUG: Undistorting image DJI_0462.JPG
2021-10-28 05:35:49,059 DEBUG: Undistorting image DJI_0463.JPG
2021-10-28 05:35:49,059 DEBUG: Undistorting image DJI_0465.JPG
2021-10-28 05:35:49,059 DEBUG: Undistorting image DJI_0468.JPG
2021-10-28 05:35:49,063 DEBUG: Undistorting image DJI_0464.JPG
2021-10-28 05:35:49,061 DEBUG: Undistorting image DJI_0467.JPG
2021-10-28 05:35:49,067 DEBUG: Undistorting image DJI_0466.JPG
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm export_visualsfm --points “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
[INFO] Finished opensfm stage
[INFO] Running openmvs stage
[INFO] running /code/SuperBuild/install/bin/opensfm/bin/opensfm export_openmvs “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm”
[INFO] Running dense reconstruction. This might take a while.
[INFO] Estimating depthmaps
[INFO] running /code/SuperBuild/install/bin/OpenMVS/DensifyPointCloud “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/undistorted/openmvs/scene.mvs” --resolution-level 2 --min-resolution 320 --max-resolution 1163 --max-threads 12 --number-views-fuse 2 -w “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/undistorted/openmvs/depthmaps” -v 0 --geometric-iters 0
05:35:53 [App ] Build date: Oct 21 2021, 01:49:08
05:35:53 [App ] CPU: Intel(R) Core™ i7-8750H CPU @ 2.20GHz (12 cores)
05:35:53 [App ] RAM: 25.00GB Physical Memory 7.00GB Virtual Memory
05:35:53 [App ] OS: Linux 5.10.16.3-microsoft-standard-WSL2 (x86_64)
05:35:53 [App ] SSE & AVX compatible CPU & OS detected
05:35:53 [App ] Command line: /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/undistorted/openmvs/scene.mvs --resolution-level 2 --min-resolution 320 --max-resolution 1163 --max-threads 12 --number-views-fuse 2 -w /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/undistorted/openmvs/depthmaps -v 0 --geometric-iters 0
05:35:53 [App ] Preparing images for dense reconstruction completed: 9 images (54ms)
05:35:53 [App ] Selecting images for dense reconstruction completed: 0 images (2ms)

Fused depth-maps 0 (100%, 0ms)
05:35:53 [App ] Densifying point-cloud completed: 0 points (59ms)
05:35:53 [App ] MEMORYINFO: {
05:35:53 [App ] VmPeak: 1652948 kB
05:35:53 [App ] VmSize: 1577448 kB
05:35:53 [App ] } ENDINFO
[INFO] running /code/SuperBuild/install/bin/OpenMVS/DensifyPointCloud --filter-point-cloud -1 -i “/var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/undistorted/openmvs/scene_dense.mvs” -v 0
05:35:53 [App ] Build date: Oct 21 2021, 01:49:08
05:35:53 [App ] CPU: Intel(R) Core™ i7-8750H CPU @ 2.20GHz (12 cores)
05:35:53 [App ] RAM: 25.00GB Physical Memory 7.00GB Virtual Memory
05:35:53 [App ] OS: Linux 5.10.16.3-microsoft-standard-WSL2 (x86_64)
05:35:53 [App ] SSE & AVX compatible CPU & OS detected
05:35:53 [App ] Command line: --filter-point-cloud -1 -i /var/www/data/f58903c5-ae4f-41f3-89ec-abe3be1877fd/opensfm/undistorted/openmvs/scene_dense.mvs -v 0
05:35:53 [App ] error: empty initial point-cloud

===== Dumping Info for Geeks (developers need this to fix bugs) =====
Child returned 1
Traceback (most recent call last):
File “/code/stages/odm_app.py”, line 94, in execute
self.first_stage.run()
File “/code/opendm/types.py”, line 347, in run
self.next_stage.run(outputs)
File “/code/opendm/types.py”, line 347, in run
self.next_stage.run(outputs)
File “/code/opendm/types.py”, line 347, in run
self.next_stage.run(outputs)
[Previous line repeated 1 more time]
File “/code/opendm/types.py”, line 328, in run
self.process(self.args, outputs)
File “/code/stages/openmvs.py”, line 161, in process
system.run(’%s %s’ % (context.omvs_densify_path, ’ '.join(config)))
File “/code/opendm/system.py”, line 106, in run
raise SubprocessException(“Child returned {}”.format(retcode), retcode)
opendm.system.SubprocessException: Child returned 1

===== Done, human-readable information to follow… =====

[ERROR] Uh oh! Processing stopped because of strange values in the reconstruction. This is often a sign that the input data has some issues or the software cannot deal with it. Have you followed best practices for data acquisition? See Flying Tips — OpenDroneMap 2.6.4 documentation

Dewdney Trunk Road - 10/27/2021

Saijin_Naib · 28 October 2021 16:12

Welcome!

Sorry you’re having trouble.

This dataset is indeed quite tricky. I think it is failing because it is not finding sufficient tiepoints to reconstruct.

Pre-processing helped it process more, but still inadequate.

Do you have the enitre dataset available?

Rob_Syvertsen · 28 October 2021 17:19

I have the full dataset on my PC at home and I’ll see if I have enough space to share it this afternoon.
At least you’ve confirmed that I should have set up a simpler test flight for my first ODM project. I’ll make sure my next test is a simple grid pattern with the camera set at -90 over something with more obvious features.

Thank you

Saijin_Naib · 28 October 2021 17:20

Mostly I think it is the motion blur combined with Rolling Shutter distortion which we currently do not correct for.

If you can set the gimbal angle, perhaps 80-85deg might be a bit better than perfectly nadir.

Rob_Syvertsen · 28 October 2021 21:50

Ok, the full dataset is posted to Google Drive now. Don’t feel obligated to process it for me, the data has no real meaning for me, it was just a throwaway test to try out ODM. I understand now that the issues are with the data capture and not things like bad EXIF data.

https://drive.google.com/drive/folders/1H4M-wPQ9zriIGzw5dKxkeZ8EWQmLFTJp?usp=sharing

system · 27 November 2021 21:51

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.