search

ultralytics / xview-yolov3

xView 2018 Object Detection Challenge: YOLOv3 Training and Inference.
https://docs.ultralytics.com
GNU Affero General Public License v3.0
252 stars 57 forks source link

Pixel coordinates #17

Closed MhmdDimassi closed 5 years ago

MhmdDimassi commented 5 years ago

Hello, thank you for this project! I am working on a project related to building detection from satellite images using YoloV3, and i think xView dataset is a big data with many labeling boxes, and this make it great for my project. But a i want to clarify some basic question related to the coordinates of Bounding Box, when part of them look negative and decimal "in geojson file" , so i think it is not a pixel coordinates, So, can you clarify this point and how to work with such coordinates or transform it to pixel coordinates (relative coordinates from top-left) ? Thank you!

glenn-jocher commented 5 years ago

@MhmdDimassi all your dataset related questions should be answered on the view dataset site: https://challenge.xviewdataset.org/welcome

MhmdDimassi commented 5 years ago

OK, thank you

sainivedh commented 1 year ago

@MhmdDimassi Hey, did you come across any info on the negative or x_max,y_max>img_width,img_height annotations ?!.