Skip to content

Tile (Slice) YOLO Dataset for Small Objects Detection

Notifications You must be signed in to change notification settings

rafcy/yolo-tiling

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

YOLO Dataset tiling script

Tile (slice) YOLO Dataset for Small Objects Detection

This script can cut images and corresponding labels from YOLO dataset into tiles of specified size and create a new datased based on these tiles. More details you can find in the article.

Usage

python3 tyle_yolo.py -source ./yolosample/ts/ -target ./yolosliced/ts/ -ext .JPG -size 512

Arguments

  • -source Source folder with images and labels needed to be tiled. Default: ./yolosample/ts/
  • -target Target folder for a new sliced dataset. Default: ./yolosliced/ts/
  • -ext Image extension in a dataset. Default: .JPG
  • -falsefolder Folder for tiles without bounding boxes
  • -size Size of a tile. Default: 416
  • -ratio Train/test split ratio. Dafault: 0.8

About

Tile (Slice) YOLO Dataset for Small Objects Detection

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%