Comments (2)
@AdeelH The second option seems more intuitive to me, and I would prefer that.
from raster-vision.
@lewfish, which of the following do you think looks more intuitive? Currently, RV is doing a mix of both which is buggy.
Option 1
Pixel indices correspond to the pixel indices in the source file. So, e.g., the first chip_sz=5
window will have a different address depending on the crop. This is simpler and closer to how RV used to handle this in earlier versions.
rs = RasterioSource(..., extent=Box(10, 10, 20, 20))
rs.extent # Box(10, 10, 20, 20)
rs.get_chip(Box(0, 0, 5, 5)) # outside the extent so blank chip
rs.get_chip(Box(10, 10, 15, 15)) # correct window to get the first chip (chip_sz=5)
rs[:5, :5] # equivalent to rs.get_chip(Box(0, 0, 5, 5)), blank chip
rs[10:15, 10:15] # equivalent to rs.get_chip(Box(10, 10, 15, 15)), first legit chip
Option 2
Here, pixel indices are relative to the crop. So the first chip_sz=5
window will always be Box(0, 0, 5, 5)
regardless of the crop, which is arguably more user-friendly.
rs = RasterioSource(..., bbox=Box(10, 10, 20, 20))
rs.extent # Box(0, 0, 10, 10), always == (0, 0, H, W)
rs.bbox # Box(10, 10, 20, 20)
rs.get_chip(Box(0, 0, 5, 5)) # correct window to get the first chip (chip_sz=5)
rs.get_chip(Box(10, 10, 15, 15)) # outside the extent so blank chip
rs[:5, :5] # equivalent to rs.get_chip(Box(0, 0, 5, 5)), first legit chip
rs[10:15, 10:15] # equivalent to rs.get_chip(Box(10, 10, 15, 15)), blank chip
from raster-vision.
Related Issues (20)
- prediction output have position shift HOT 13
- Some tiles get incorrect when predicting but of 95% are ok HOT 9
- Support reading temporal data (i.e. time-series of multiple images of the same scene)
- Can't use a geojson as AOI in a SemanticSegmentationRandomWindowGeoDataset HOT 3
- ARM64 build currently broken HOT 1
- Unable to install RasterVision HOT 3
- Issues with using model bundle for prediction HOT 15
- Cannot import ClassConfig on Kaggle HOT 16
- Cannot save prediction using colors from ClassConfig HOT 4
- Improve unit test coverage of CLI and `Runner`s
- Cannot plot batch with ObjectDetectionVisualizer HOT 4
- Multi-temporal raster source visualizer fails when batch size is 1 HOT 2
- Make it possible to exclude "null" class labels from the computation of metrics HOT 3
- RuntimeError: expected scalar type Long but found Int HOT 10
- Allow user to specify AOI box filtering behavior in sliding window datasets HOT 1
- self._hds cannot be converted to a Python object for pickling HOT 2
- Semantic Segmentation Labels not initializing properly from predictions when extent provided HOT 2
- use my trained modle to prediction ,has wrong happened HOT 2
- RuntimeError: The size of tensor a (82) must match the size of tensor b (64) at non-singleton dimension 3 HOT 4
- Migrate to `pydantic` v2
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 raster-vision.