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

Domain shift degrades the performance of object detection models in practicalapplications. To alleviate the influence of domain shift, plenty of previouswork try to decouple and learn the domain-invariant (common) features fromsource domains via domain adversarial learning (DAL). However, inspired bycausal mechanisms, we find that previous methods ignore the implicitinsignificant non-causal factors hidden in the common features. This is mainlydue to the single-view nature of DAL. In this work, we present an idea toremove non-causal factors from common features by multi-view adversarialtraining on source domains, because we observe that such insignificantnon-causal factors may still be significant in other latent spaces (views) dueto the multi-mode structure of data. To summarize, we propose a Multi-viewAdversarial Discriminator (MAD) based domain generalization model, consistingof a Spurious Correlations Generator (SCG) that increases the diversity ofsource domain by random augmentation and a Multi-View Domain Classifier (MVDC)that maps features to multiple latent spaces, such that the non-causal factorsare removed and the domain-invariant features are purified. Extensiveexperiments on six benchmarks show our MAD obtains state-of-the-artperformance.

Multi-view Adversarial Discriminator: Mine the Non-causal Factors for Object Detection in Unseen Domains