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

Large-scale well-annotated datasets are of great importance for training aneffective object detector. However, obtaining accurate bounding box annotationsis laborious and demanding. Unfortunately, the resultant noisy bounding boxescould cause corrupt supervision signals and thus diminish detectionperformance. Motivated by the observation that the real ground-truth is usuallysituated in the aggregation region of the proposals assigned to a noisyground-truth, we propose DIStribution-aware CalibratiOn (DISCO) to model thespatial distribution of proposals for calibrating supervision signals. InDISCO, spatial distribution modeling is performed to statistically extract thepotential locations of objects. Based on the modeled distribution, threedistribution-aware techniques, i.e., distribution-aware proposal augmentation(DA-Aug), distribution-aware box refinement (DA-Ref), and distribution-awareconfidence estimation (DA-Est), are developed to improve classification,localization, and interpretability, respectively. Extensive experiments onlarge-scale noisy image datasets (i.e., Pascal VOC and MS-COCO) demonstratethat DISCO can achieve state-of-the-art detection performance, especially athigh noise levels.

Distribution-Aware Calibration for Object Detection with Noisy Bounding Boxes