Humble Teachers Teach Better Students for Semi-Supervised Object Detection
This work addresses the problem of reducing annotation costs for object detection in computer vision, offering a semi-supervised approach that is incremental but provides strong performance gains.
The paper tackles semi-supervised object detection by proposing a teacher-student framework with soft pseudo-labels and data ensemble, achieving a COCO-style AP of 53.04% on VOC07 val set, which is 8.4% better than the prior state-of-the-art STAC, and reaching 53.8% AP on MS-COCO test-dev with a 3.1% gain over fully supervised methods.
We propose a semi-supervised approach for contemporary object detectors following the teacher-student dual model framework. Our method is featured with 1) the exponential moving averaging strategy to update the teacher from the student online, 2) using plenty of region proposals and soft pseudo-labels as the student's training targets, and 3) a light-weighted detection-specific data ensemble for the teacher to generate more reliable pseudo-labels. Compared to the recent state-of-the-art -- STAC, which uses hard labels on sparsely selected hard pseudo samples, the teacher in our model exposes richer information to the student with soft-labels on many proposals. Our model achieves COCO-style AP of 53.04% on VOC07 val set, 8.4% better than STAC, when using VOC12 as unlabeled data. On MS-COCO, it outperforms prior work when only a small percentage of data is taken as labeled. It also reaches 53.8% AP on MS-COCO test-dev with 3.1% gain over the fully supervised ResNet-152 Cascaded R-CNN, by tapping into unlabeled data of a similar size to the labeled data.