Comments (13)
Thanks @simonjchapman for pointing this out. I'll follow up unce we track this down.
from usbuildingfootprints.
FWIW, this is a problem in Chicago, too.
from usbuildingfootprints.
+1 on this, would love it (if possible) for the missing tiles to be included. This data is great to have, thanks again!
from usbuildingfootprints.
There are also tile-shaped gaps all along I-15 in Connecticut.
from usbuildingfootprints.
I've looked at several states, it appears to be any major metropolitan area has these gaps. I'd imagine it is something in the workflow that will be fixed for all when one is fixed.
from usbuildingfootprints.
The gap areas contain image tiles taken with different cameras, which is causing the creation of artificial edges between neighboring tiles. These confuse our detection network which hasn't learned to deal with them.
We took a very conservative approach of skipping such tiles. I think we could add aditional effort in order to properly deal with this problem.
from usbuildingfootprints.
Very interesting. Might it be good to relax your approach and output these tiles as a "delta"? That would also allow members of this group to help you with validating their quality and also not disrupt the integrity of the currently published set.
Thanks again for your efforts in this area.
from usbuildingfootprints.
Fixed in the latest release.
from usbuildingfootprints.
It looks like this may still be happening in a few spots in the 1.1 release. I noticed it in the Texas, California, and Colorado data:
Around downtown Houston:
In L.A., south of the Hollywood Hills:
Some more hard-to-see artifacts around San Jose:
And around downtown Denver:
from usbuildingfootprints.
@simonjchapman How have you generated your renderings? I'm looking for the right software ...
from usbuildingfootprints.
In at least the San Jose example, it looks like it is being too conservative for the edge of East Capitol Freeway:
looks like it tracks to the 101 just fine but there's a nice hard edge on the imagery (used by Apple Maps)
from usbuildingfootprints.
Great work pulling this all together. Just wanted to make you aware that we're seeing similar issues near Minneapolis/St. Paul.
from usbuildingfootprints.
Incredible dataset! Likewise, Louisville Kentucky.
from usbuildingfootprints.
Related Issues (20)
- 2018 buildings disappearing from 2021 edition
- Question: What % of all US buildings are estimated to be covered? HOT 1
- More details about polygonization algorithm? HOT 2
- Footprints for Ukraine HOT 1
- Extract polygons for a specific city from State GeoJson HOT 1
- which version of EfficientNet?
- Alternative to GeoJSON? HOT 5
- FootPrint of Nepal
- Satellite Image of Kathmandu in 2015 (Pre and Post Earthquake)
- Alaska missing a lot of areas/buildings?
- Unable To Download Data HOT 3
- Update has fewer buildings for US
- Footprints less accurate than they were before. HOT 2
- Wrong shape files showing up for Aberdeen Proving Grounds, Harford County, Maryland, US
- Tall buildings not showing up correctly in Chicago HOT 1
- This repo is missing a LICENSE file
- This repo is missing important files HOT 1
- Delta in QGIS
- I need to get my Microsoft points back. Help me please. HOT 1
- Building Heights HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from usbuildingfootprints.