search

MingtaoFu / gliding_vertex

The implementation of paper "Gliding vertex on the horizontal bounding box for multi-oriented object detection".
271 stars 63 forks source link

If I want to use this model to train my own data set, how should I label the data? #1

Closed NuaaWill closed 4 years ago

MingtaoFu commented 4 years ago

DOTA's default annotation format is *.txt, which is composed of text lines: x1, y1, x2, y2, x3, y3, x4, y4, cls, diff. In this project, we convert the txt annotations to json annotations with txt2json.py. I updated the README.md and added the section If you want to train with your own data in it. Please refer to it.

MingtaoFu commented 4 years ago

If you need more details, feel free to make additional comments.