English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

This paper focuses on the scale imbalance problem of semi-supervised objectdetection(SSOD) in aerial images. Compared to natural images, objects in aerialimages show smaller sizes and larger quantities per image, increasing thedifficulty of manual annotation. Meanwhile, the advanced SSOD technique cantrain superior detectors by leveraging limited labeled data and massiveunlabeled data, saving annotation costs. However, as an understudied task inaerial images, SSOD suffers from a drastic performance drop when facing a largeproportion of small objects. By analyzing the predictions between small andlarge objects, we identify three imbalance issues caused by the scale bias,i.e., pseudo-label imbalance, label assignment imbalance, and negative learningimbalance. To tackle these issues, we propose a novel Scale-discriminativeSemi-Supervised Object Detection (S^3OD) learning pipeline for aerial images.In our S^3OD, three key components, Size-aware Adaptive Thresholding (SAT),Size-rebalanced Label Assignment (SLA), and Teacher-guided Negative Learning(TNL), are proposed to warrant scale unbiased learning. Specifically, SATadaptively selects appropriate thresholds to filter pseudo-labels for objectsat different scales. SLA balances positive samples of objects at differentscales through resampling and reweighting. TNL alleviates the imbalance innegative samples by leveraging information generated by a teacher model.Extensive experiments conducted on the DOTA-v1.5 benchmark demonstrate thesuperiority of our proposed methods over state-of-the-art competitors. Codeswill be released soon.

Rethinking Scale Imbalance in Semi-supervised Object Detection for Aerial Images