what is the "match type " and "keypoints" in the ic15 datasets?

wangyuxin87 / ContourNet

A PyTorch implementation of "ContourNet: Taking a Further Step toward Accurate Arbitrary-shaped Scene Text Detection" (CVPR2020)

Other

227 stars 46 forks source link

what is the "match type " and "keypoints" in the ic15 datasets? #10

Closed qiaolegeke closed 4 years ago

qiaolegeke commented 4 years ago

I try to train my own dataset ,but i don't know this 2 value,

if i need to modifiy somewhere?

when i { "info": {}, "license": [], "images": [ { "height": 1422, "width": 1062, "license": 0, "id": 0, "file_name": "img_1.jpg" }, .... "annotations": [ { "image_id": 0, "segmentation": [ [ 956, 624, 1060, 624, 1060, 662, 956, 662 ] ], "bbox": [ 956, 624, 105, 39 ], "area": 3952.0, "category_id": 1, "id": 0, "iscrowd": 0 },

this is my json