Consistent Teacher Provides Better Supervision in Semi-supervised Object Detection

by   Xinjiang Wang, et al.
SenseTime Corporation

In this study, we dive deep into the unique challenges in semi-supervised object detection (SSOD). We observe that current detectors generally suffer from 3 inconsistency problems. 1) Assignment inconsistency, that the conventional assignment policy is sensitive to labeling noise. 2) Subtasks inconsistency, where the classification and regression predictions are misaligned at the same feature point. 3) Temporal inconsistency, that the pseudo bboxes vary dramatically at different training steps. These issues lead to inconsistent optimization objectives of the student network, thus deteriorating performance and slowing down the model convergence. We, therefore, propose a systematic solution, termed Consistent Teacher, to remedy the above-mentioned challenges. First, adaptive anchor assignment substitutes the static IoU-based strategy, which enables the student network to be resistant to noisy psudo bboxes; Then we calibrate the subtask predictions by designing a feature alignment module; Lastly, We adopt a Gaussian Mixture Model (GMM) to dynamically adjust the pseudo-boxes threshold. Consistent Teacher provides a new strong baseline on a large range of SSOD evaluations. It achieves 40.0 mAP with ResNet-50 backbone given only 10 data, which surpasses previous baselines using pseudo labels by around 4 mAP. When trained on fully annotated MS-COCO with additional unlabeled data, the performance further increases to 49.1 mAP. Our code will be open-sourced soon.


page 22

page 23

page 24


Humble Teachers Teach Better Students for Semi-Supervised Object Detection

We propose a semi-supervised approach for contemporary object detectors ...

Unbiased Teacher v2: Semi-supervised Object Detection for Anchor-free and Anchor-based Detectors

With the recent development of Semi-Supervised Object Detection (SS-OD) ...

Adaptive Self-Training for Object Detection

Deep learning has emerged as an effective solution for solving the task ...

Label Matching Semi-Supervised Object Detection

Semi-supervised object detection has made significant progress with the ...

Hierarchical Supervision and Shuffle Data Augmentation for 3D Semi-Supervised Object Detection

State-of-the-art 3D object detectors are usually trained on large-scale ...

Learning Domain Adaptive Object Detection with Probabilistic Teacher

Self-training for unsupervised domain adaptive object detection is a cha...

Please sign up or login with your details

Forgot password? Click here to reset