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

Nearly all existing scene graph generation (SGG) models have overlooked theground-truth annotation qualities of mainstream SGG datasets, i.e., theyassume: 1) all the manually annotated positive samples are equally correct; 2)all the un-annotated negative samples are absolutely background. In this paper,we argue that neither of the assumptions applies to SGG: there are numerousnoisy ground-truth predicate labels that break these two assumptions and harmthe training of unbiased SGG models. To this end, we propose a novel NoIsylabel CorrEction and Sample Training strategy for SGG: NICEST. Specifically, itconsists of two parts: NICE and NIST, which rule out these noisy label issuesby generating high-quality samples and the effective training strategy,respectively. NICE first detects noisy samples and then reassigns them morehigh-quality soft predicate labels. NIST is a multi-teacher knowledgedistillation based training strategy, which enables the model to learn unbiasedfusion knowledge. And a dynamic trade-off weighting strategy in NIST isdesigned to penalize the bias of different teachers. Due to the model-agnosticnature of both NICE and NIST, our NICEST can be seamlessly incorporated intoany SGG architecture to boost its performance on different predicatecategories. In addition, to better evaluate the generalization of SGG models,we further propose a new benchmark VG-OOD, by re-organizing the prevalent VGdataset and deliberately making the predicate distributions of the training andtest sets as different as possible for each subject-object category pair. Thisnew benchmark helps disentangle the influence of subject-object category basedfrequency biases. Extensive ablations and results on different backbones andtasks have attested to the effectiveness and generalization ability of eachcomponent of NICEST.

NICEST: Noisy Label Correction and Training for Robust Scene Graph Generation