Skip to main content
. 2024 Aug 20;15:1396568. doi: 10.3389/fpls.2024.1396568

Table 1.

Testing performance (mAP@[0.5:0.95]) comparison between the supervised and semi-supervised based on Faster-RCNN and FCOS models on the CottonWeedDet3 dataset.

Algorithms Supervision type Proportion of labeled data for training
5% 10% 20% 50% 100%
Faster-RCNN Supervised 21.14 31.90 42.65 50.51 56.75
Semi-supervised 29.33 40.14 49.20 53.19
FCOS Supervised 27.37 42.20 52.42 59.84 62.80
Semi-supervised 38.17 47.93 55.79 61.32