mapillary / mapillary/OpenSfM

Poor Quality Point Cloud in SfM with Custom Configuration

Open
#1,058 0 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
3.8k
Forks
899
PR merge metrics
No merged PRs in 30d

Description

I am experiencing issues with the quality of the point cloud generated by my Sfm pipeline. Despite trying various configuration parameters, the resulting point cloud is not good enough. Below are the details of my setup and the steps I have taken so far.

Configuration:
depthmap_min_consistent_views: 3
hahog_edge_threshold: 30
depthmap_save_debug_files: 'no'
feature_min_frames: 12000
Steps Taken:

Captured several images with a distance of 5 meters between each image.
Added an exif_overrides file to the project.
Experimented with various configuration parameters, but the point cloud quality remains unsatisfactory.
Images and Results:
pc3
pc2
pc1

Questions:

Are there any issues with my current configuration that could be causing poor point cloud quality?
What other configuration parameters should I try to improve the point cloud?
Are there best practices for capturing images or additional preprocessing steps that I should follow?
Additional Information:

Attachments:
sampleimage

Sample images of the point cloud
Thank you for your assistance!
@fabianschenk @YanNoun @paulinus

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No source file, test, or entry point is named. Start by reproducing the SfM pipeline with the listed settings and sample images, then inspect the configuration and reconstruction diagnostics. Done means identifying a reproducible cause or documenting validated capture and configuration guidance.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.