Missing Features


I’m new(-ish) to WebODM, and I’m having some issues. The image below is of a quarry that I’ve been doing some work in, and as you can see, there are sections of what appears to be missing data. These are bunds (soil/rock mounds placed on top of a quarry face to act as a barrier) … I’ve tried adjusting parameters to get the bunds to show, and run it on high too (225M points), but each time it’s as if WebODM thinks they’re an error and is stripping them out. What can I tweak to stop that from happening ? I’ve had a cloud back from Dronedeploy and that shows the bunds, so I’m sure it’s not an image density issue. Any help would be much appreciated.

Thanks in advance.



Can you please put your full processing parameters here so we can get a better idea of what you are running?

Hi Saijin,

Thanks for responding so quickly. Please find the parameters below … I’ve taken this from the taskoutput.txt file …

[INFO] Initializing ODM - Mon Mar 15 19:21:01 2021
[INFO] ==============
[INFO] build_overviews: False
[INFO] camera_lens: auto
[INFO] cameras: {}
[INFO] crop: 3
[INFO] debug: False
[INFO] dem_decimation: 1
[INFO] dem_euclidean_map: False
[INFO] dem_gapfill_steps: 8
[INFO] dem_resolution: 3.0
[INFO] depthmap_resolution: 1200.0
[INFO] dsm: True
[INFO] dtm: False
[INFO] end_with: odm_report
[INFO] fast_orthophoto: False
[INFO] feature_quality: medium
[INFO] feature_type: sift
[INFO] force_gps: False
[INFO] gcp: /var/www/data/d096b501-f868-4381-925e-f66582300729/gcp/gcp_list.txt
[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: 6
[INFO] merge: all
[INFO] mesh_octree_depth: 12
[INFO] mesh_size: 200000
[INFO] min_num_features: 8000
[INFO] name: d096b501-f868-4381-925e-f66582300729
[INFO] optimize_disk_space: False
[INFO] orthophoto_compression: DEFLATE
[INFO] orthophoto_cutline: 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: False
[INFO] pc_filter: 0.0
[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_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: True
[INFO] use_exif: False
[INFO] use_fixed_camera_params: False
[INFO] use_hybrid_bundle_adjustment: False
[INFO] verbose: False
[INFO] ==============
[INFO] Running dataset stage
[INFO] Loading dataset from: /var/www/data/d096b501-f868-4381-925e-f66582300729/images
[INFO] Loading 916 images
[INFO] Wrote images database: /var/www/data/d096b501-f868-4381-925e-f66582300729/images.json
[INFO] Found 916 usable images
[INFO] Parsing SRS header: WGS84 UTM 30N
[INFO] Parsing SRS header: WGS84 UTM 30N
[INFO] Parsing SRS header: WGS84 UTM 30N
[INFO] 24 GCP points will be used for georeferencing
[INFO] Generated coords file from GCP: WGS84 UTM 30N
[INFO] Wrote /var/www/data/d096b501-f868-4381-925e-f66582300729/odm_georeferencing/odm_georeferencing_model_geo.txt
[INFO] Parsing SRS header: WGS84 UTM 30N
[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: 4000px
[WARNING] Legacy option --depthmap-resolution (this might be removed in a future version). Use --pc-quality instead.
[INFO] Altitude data detected, enabling it for GPS alignment
[INFO] [‘use_exif_size: no’, ‘flann_algorithm: KDTREE’, ‘feature_process_size: 1000’, ‘feature_min_frames: 8000’, ‘processes: 6’, ‘matching_gps_neighbors: 8’, ‘matching_gps_distance: 0’, ‘optimize_camera_parameters: yes’, ‘undistorted_image_format: tif’, ‘bundle_outlier_filtering_type: AUTO’, ‘align_orientation_prior: vertical’, ‘triangulation_type: ROBUST’, ‘retriangulation_ratio: 2’, ‘feature_type: SIFT’, ‘use_altitude_tag: yes’, ‘align_method: auto’, ‘local_bundle_radius: 0’, ‘bundle_use_gcp: yes’, ‘bundle_use_gps: no’]
[INFO] running /code/SuperBuild/src/opensfm/bin/opensfm extract_metadata “/var/www/data/d096b501-f868-4381-925e-f66582300729/opensfm”

00:04:35 [App ] Densifying point-cloud completed: 273376931 points (3h39m30s794ms)
00:14:11 [App ] MEMORYINFO: {
00:14:11 [App ] VmPeak: 84096872 kB
00:14:11 [App ] VmSize: 79620216 kB
00:14:11 [App ] } ENDINFO
[INFO] running /code/SuperBuild/install/bin/OpenMVS/DensifyPointCloud --filter-point-cloud -1 -i “/var/www/data/d096b501-f868-4381-925e-f66582300729/opensfm/undistorted/openmvs/scene_dense.mvs” -v 0
00:14:31 [App ] Build date: Mar 7 2021, 19:02:49
00:14:31 [App ] CPU: AMD Ryzen 9 3900X 12-Core Processor (6 cores)
00:14:31 [App ] RAM: 117.94GB Physical Memory 30.00GB Virtual Memory
00:14:31 [App ] OS: Linux 5.4.72-microsoft-standard-WSL2 (x86_64)
00:14:31 [App ] SSE & AVX compatible CPU & OS detected
00:14:31 [App ] Command line: --filter-point-cloud -1 -i /var/www/data/d096b501-f868-4381-925e-f66582300729/opensfm/undistorted/openmvs/scene_dense.mvs -v 0

1 Like

Let’s try a few tweaks:
crop 0
feature_quality: ultra
pc_quality: ultra
resize_to: -1
use_hybrid_bundle_adjustment: true

A post was split to a new topic: Missing Features during Reconstruction

Many thanks Saijin, I’ll apply those and remodel. I’ll let you know whether it works.

1 Like

Hi, me again. :slight_smile:

I’ve ran the ultra version with the parameters you’ve suggested, and the same has occurred. I’ve attached two screenshots for you to see. One shows the missing ‘bund’ again, the second shows a pipe lying on the ground, which has been picked up well. It seems that the bund is being decimated / filtered out somehow. Any ideas ? Thanks again for your help.Pipe Ultra

1 Like

Looks like the pros might need to step in. Anything else from me at this point will just be shots in the dark :slight_smile:

1 Like

Thanks for your help so far Saijin. That Ultra took 19hrs and 32 mins btw (!!) … To me, it’s almost as if WebODM thinks the bund is part of the DSM, and it wants to output a DTM instead. I’ll wait to hear from the pros … hopefully one will appear in a puff of smoke soon. :slight_smile:

1 Like

This is the difference between the WebODM model (Heightbanded), and the Drone Deploy version (RGB). Just another way of looking at the two datasets and the parts that are missing.image3

1 Like