🚀 Excited to share a new tutorial on how to detect buildings 🏘️using the Segment Anything Model 2…
Qiusheng Wu
在 X 上查看@giswqs5.1万 位关注者
72,345 次观看
🚀 Excited to share a new tutorial on how to detect buildings 🏘️using the Segment Anything Model 2 (SAM 2) with Segment-Geospatial! You can choose interactive segmentation or utilize a GeoJSON file with building centroids or bounding boxes to do automated building detection.
📹 Check out the video tutorial:
📓 Explore the notebook:
#Geospatial #OpenSource #AI #MachineLearning
10 条评论
评论来自 X (Twitter) 上的原帖

@disaster_db
Amazing how things have improved with image segmentation since eCognition came out
2 个赞

@thegerritcloete
This is amazing work! Keep it up!
1 个赞

@apolmig
thanks! always awesome content and learning
1 个赞

@aleix10kst
@ograuc
1 个赞

@neurallambda
@ChiasticChaos This is exactly(ish?) what you were asking about!

@1dolinski
looks great -- what's your favourite use case so far?

@sinsofsid
Can you model Square Footage of the buildings/locations with 80%+ accuracy? Use case: Input location address and output SQFT?

@nahkhsinad
This is awesome

@jonathanalis1
Can I just bound box the full image, and it will get all the building masks?

@giswqs
No. See this bbox example.