microsoft / microsoft/GlobalMLBuildingFootprints

Strange Shapes Generated

Open
#129 6 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
2k
Forks
277
PR merge metrics
No merged PRs in 30d

Description

Firstly, this is a great set of data that I'm happy to have access to - thank you.

I've been validating some of the data and have seen some strange polygons created and wanted to know what the root cause of this is (whether I'm doing something wrong, etc)

Take for example the Lee Valley Velo Park in the London Olympic Village

QuadKey
031313131

GeoJSON

{"type": "Feature", "properties": { "height": 13.368171691894531, "confidence": -1.0},"geometry": {"type": "Polygon","coordinates": [[[-0.0157944811891353, 51.5498873613391], [-0.0150260252882228, 51.54981204486004], [-0.01432737688843, 51.55010193634923], [-0.0142696764511419, 51.55032957855235], [-0.0143002375035706, 51.550471440745774], [-0.014356194905103, 51.55047692506388], [-0.0143442665252391, 51.55048187447204], [-0.014487055323984, 51.55046997412536], [-0.0144380864679139, 51.55053708350929], [-0.0144016262879187, 51.550503106682434], [-0.0143764661721479, 51.550513546289885], [-0.0143894566669678, 51.550517212058246], [-0.0142963400004903, 51.55052496855504], [-0.0146877720281213, 51.55088973810615], [-0.0147126724344382, 51.55087940634456], [-0.0147352944973411, 51.55098441509665], [-0.015145624364965, 51.551024630568165], [-0.0154314963522073, 51.551000818062036], [-0.0154610816810541, 51.55088415044336], [-0.0155460087835846, 51.55087707248641], [-0.015393999752602, 51.5508621743923], [-0.0154913992232708, 51.5508540612036], [-0.0152974538954131, 51.55079934087585], [-0.0157401026437887, 51.55084272402034], [-0.0157342952068773, 51.55081576657256], [-0.0158801282427845, 51.55075525689982], [-0.0158274894931654, 51.55074040521631], [-0.0158615216231595, 51.55072628444183], [-0.0159005322309236, 51.55076263780404], [-0.0162377862542717, 51.55030051949016], [-0.0157944811891353, 51.5498873613391]]]}}

However, this yields

Image

Also, in other more simple shapes, the polygon regularly seems off-center, e.g. the following GeoJSON from the same quadkey

{"type": "Feature", "properties": { "height": 7.860638618469238, "confidence": -1.0},"geometry": {"type": "Polygon","coordinates": [[[-0.0274372725260852, 51.5481362747604], [-0.0272214088040679, 51.547928701888935], [-0.0268442539518987, 51.54808039258931], [-0.027034435718174, 51.54826324954267], [-0.027223555193634, 51.5481871947816], [-0.0272492415839887, 51.54821189185068], [-0.0274372725260852, 51.5481362747604]]]}}

Yields

Image

Any help would be appreciated, thanks in advance

Contributor guide

No contributing guide indexed for this repository

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

Start by reproducing the Lee Valley Velo Park examples using QuadKey 031313131, the supplied GeoJSON, and the attached images. Compare the generated polygons with the source data to determine why shapes are distorted or off-center; done means identifying the root cause and documenting or correcting the affected output.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.