Method: TH-DL - Samples list - Task 3 - Joint text detection and script identification - ICDAR2017 Competition on Multi-lingual scene text detection and script identification - Robust Reading Competition

Page of 450

next >

View

Thumbnails
Table

Sample	Results
	Global
	Average Precision	Hmean	Precision	Recall
1	66.67%	72.73%	80.00%	66.67%
2	14.88%	25.00%	50.00%	16.67%
3	0.00%	0.00%	0.00%	0.00%
4	0.00%	0.00%	0.00%	0.00%
5	33.33%	44.44%	66.67%	33.33%
6	66.67%	50.00%	40.00%	66.67%
7	25.00%	15.38%	11.11%	25.00%
8	43.94%	50.00%	55.56%	45.45%
9	6.22%	11.36%	26.32%	7.25%
10	0.00%	0.00%	0.00%	0.00%
11	16.67%	36.36%	40.00%	33.33%
12	52.33%	57.14%	54.55%	60.00%
13	30.21%	48.00%	60.00%	40.00%
14	1.18%	8.70%	16.67%	5.88%
15	100.00%	100.00%	100.00%	100.00%
16	19.40%	40.00%	66.67%	28.57%
17	12.50%	28.57%	33.33%	25.00%
18	50.00%	66.67%	100.00%	50.00%
19	50.00%	66.67%	100.00%	50.00%
20	71.43%	76.92%	83.33%	71.43%

Download full table (.csv)

Results