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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Object detection in aerial images is a pivotal task for various earthobservation applications, whereas current algorithms learn to detect only apre-defined set of object categories demanding sufficient bounding-boxannotated training samples and fail to detect novel object categories. In thispaper, we consider open-vocabulary object detection (OVD) in aerial images thatenables the characterization of new objects beyond training categories on theearth surface without annotating training images for these new categories. Theperformance of OVD depends on the quality of class-agnostic region proposalsand pseudo-labels that can generalize well to novel object categories. Tosimultaneously generate high-quality proposals and pseudo-labels, we proposeCastDet, a CLIP-activated student-teacher open-vocabulary object Detectionframework. Our end-to-end framework within the student-teacher mechanismemploys the CLIP model as an extra omniscient teacher of rich knowledge intothe student-teacher self-learning process. By doing so, our approach boostsnovel object proposals and classification. Furthermore, we design a dynamiclabel queue technique to maintain high-quality pseudo labels during batchtraining and mitigate label imbalance. We conduct extensive experiments onmultiple existing aerial object detection datasets, which are set up for theOVD task. Experimental results demonstrate our CastDet achieving superioropen-vocabulary detection performance, e.g., reaching 40.0 HM (Harmonic Mean),which outperforms previous methods Detic/ViLD by 26.9/21.1 on the VisDroneZSDdataset.

CastDet: Toward Open Vocabulary Aerial Object Detection with CLIP-Activated Student-Teacher Learning